Advances in MS-Based Analytical Methods: Innovations and Future Trends

Journal of Analytical Methods in Chemistry

Lead Guest Editor: Federica Bianchi Guest Editors: Leopold Ilag, Veronica Termopoli, and Lucia Mendez Advances in MS-Based Analytical Methods: Innovations and Future Trends Journal of Analytical Methods in Chemistry

Advances in MS-Based Analytical Methods: Innovations and Future Trends

This is a special issue published in “Journal of Analytical Methods in Chemistry.” All articles are open access articles distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Editorial Board

Mohamed Abdel-Rehim, Sweden Karoly Heberger, Hungary Pablo Richter, Chile Hassan Y. Aboul Enein, Egypt A. V. Herrera-Herrera, Spain Fábio Rodrigo Piovezan Rocha, Brazil Silvana Andreescu, USA Eliseo Herrero-Hernández, Spain Erwin Rosenberg, Austria Aristidis N. Anthemidis, Greece Bernd Hitzmann, Germany Jose Vicente Ros-Lis, Spain Alessandro Buccolieri, Italy Chih-Ching Huang, Taiwan Giuseppe Ruberto, Italy Antony C. Calokerinos, Greece Jaroon Jakmunee, Thailand Antonio Ruiz Medina, Spain Luca Campone, Italy Christos Kontoyannis, Greece Bradley B. Schneider, Canada Ricardo Jorgensen Cassella, Brazil Radosław Kowalski, Poland Jesus Simal-Gandara, Spain Angela Chambery, Italy Eulogio J. Llorent-Martínez, Spain Hana Sklenarova, Czech Republic Igor Chourpa, France Mercedes G. Lopez, Mexico Beate Strehlitz, Germany Filomena Conforti, Italy Miren Lopez de Alda, Spain Luca Tortora, Italy Guido Crisponi, Italy Larisa Lvova, Italy Marek Trojanowicz, Poland Eduardo Dellacassa, Uruguay Jose Carlos Marques, Portugal Bengi Uslu, Turkey Ana María Díez-Pascual, Spain Christophe A. Marquette, France Anna Vallverdu-Queralt, France Gauthier Eppe, Belgium Serban C. Moldoveanu, USA Krishna K. Verma, India Josep Esteve-Romero, Spain Yolanda Moliner Martínez, Spain Brian K. Via, USA Valdemar Esteves, Portugal Paolo Montuori, Italy Adam Voelkel, Poland Núria Fontanals, Spain SibelA.Ozkan,Turkey Rongda Xu, USA Constantinos Georgiou, Greece Federica Pellati, Italy B. M. Nikolova-Damyanova, Bulgaria Gabriele Giancane, Italy Verónica Pino, Spain Contents

Advances in MS-Based Analytical Methods: Innovations and Future Trends Federica Bianchi ,LeopoldIlag , Veronica Termopoli ,andLuciaMendez Editorial (2 pages), Article ID 2084567, Volume 2018 (2018)

New Advances in Toxicological Forensic Analysis Using Mass Spectrometry Techniques Noroska Gabriela Salazar Mogollón , Cristian Daniel Quiroz-Moreno, Paloma Santana Prata, Jose Rafael de Almeida, Amanda Sofía Cevallos, Roldán Torres-Guiérrez, and Fabio Augusto Review Article (17 pages), Article ID 4142527, Volume 2018 (2018)

Establishing Analytical Performance Criteria for the Global Reconnaissance of Antibiotics and Other Pharmaceutical Residues in the Aquatic Environment Using Liquid Chromatography-Tandem Mass Spectrometry Luisa F. Angeles and Diana S. Aga Research Article (9 pages), Article ID 7019204, Volume 2018 (2018)

Determination of Tobramycin in M9 Medium by LC-MS/MS: Signal Enhancement by Trichloroacetic Acid Liusheng Huang , Janus Anders Juul Haagensen, Davide Verotta , Vincent Cheah, Alfred M. Spormann, Francesca Aweeka, and Katherine Yang Research Article (8 pages), Article ID 7965124, Volume 2018 (2018)

MS-Based Analytical Techniques: Advances in Spray-Based Methods and EI-LC-MS Applications Federica Bianchi , Nicolò Riboni , Veronica Termopoli ,LuciaMendez ,IsabelMedina, Leopold Ilag , Achille Cappiello, and Maria Careri Review Article (24 pages), Article ID 1308167, Volume 2018 (2018)

87 86 High-Precision In Situ Sr/ Sr Analyses through Microsampling on Solid Samples: Applications to Earth and Life Sciences Sara Di Salvo, Eleonora Braschi , Martina Casalini, Sara Marchionni, Teresa Adani, Maurizio Ulivi, Andrea Orlando ,SimoneTommasini,RiccardoAvanzinelli,PaulP.A.Mazza,SandroConticelli , and Lorella Francalanci Research Article (20 pages), Article ID 1292954, Volume 2018 (2018)

Development and Validation of an LC-MS/MS Method and Comparison with a GC-MS Method to Measure Phenytoin in Human Brain Dialysate, Blood, and Saliva Raphael Hösli ,StefanKönig,andStefanF.Mühlebach Research Article (8 pages), Article ID 8274131, Volume 2018 (2018)

Analysis of Polycyclic Aromatic Hydrocarbons in Ambient Aerosols by Using One-Dimensional and Comprehensive Two-Dimensional Gas Chromatography Combined with Mass Spectrometric Method: AComparativeStudy Yun Gyong Ahn , So Hyeon Jeon, Hyung Bae Lim, Na Rae Choi, Geum-Sook Hwang, Yong Pyo Kim, and Ji Yi Lee Research Article (9 pages), Article ID 8341630, Volume 2018 (2018) Hindawi Journal of Analytical Methods in Chemistry Volume 2018, Article ID 2084567, 2 pages https://doi.org/10.1155/2018/2084567

Editorial Advances in MS-Based Analytical Methods: Innovations and Future Trends

Federica Bianchi ,1 Leopold Ilag ,2 Veronica Termopoli ,3 and Lucia Mendez 4

1Department of Chemistry, Life Sciences and Environmental Sustainability, Parma University, Parco Area delle Scienze 17/A, 43124 Parma, Italy 2Department of Environmental Science and Analytical Chemistry, Stockholm University, 10691 Stockholm, Sweden 3Department of Pure and Applied Sciences, LC-MS Laboratory, Piazza Rinascimento 6, 61029 Urbino, Italy 4Instituto de Investigaciones Marinas, Spanish National Research Council (IIM-CSIC), Eduardo Cabello, 6, E-36208 Vigo, Spain

Correspondence should be addressed to Federica Bianchi; [email protected]

Received 9 September 2018; Accepted 10 September 2018; Published 8 October 2018

Copyright © 2018 Federica Bianchi et al. $is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Mass spectrometry (MS) is a widely used technique for food $is special issue covers the broad area of MS-based safety, environmental, pharmaceutical, biological, and fo- analytical methods starting from the development of novel rensic investigations where the simultaneous detection of LC-MS and LC-MS/MS methods for the quantitation of targeted and nontargeted compounds is of pivotal impor- compounds of environmental and pharmaceutical concern tance. A plethora of analytical MS methods also coupled to to the use of more advanced separation technologies coupled different separation techniques such as gas- and liquid- to mass spectrometry for the analysis of atmospheric chromatography and their multidimensional analogues or samples. More precisely, F. Bianchi et al. have provided capillary electrophoresis have been developed and validated a discussion of some instrumental innovations and their in order to analyze complex matrices. However, despite the applications in the field of mass spectrometry with particular rapid evolution from its beginning, the development of attention to spray-based MS methods and LC-EI-based MS online and real-time analytical MS methodologies especially interfaces. New materials, prototypes, and instrumental ambient ionization methods is strongly demanded to per- configurations able to increase the performance of the de- form high-throughput analysis and to obtain highly in- veloped methods are presented and discussed. Finally, an formative spectra. In this context, novel materials and overview of the most recent MS-based methods in food instrumental configurations are under study to enhance the analysis is given covering the state of the art from 2012 up to performance of the different instruments, whereas powerful 2017. L. F. Angeles and D. S. Aga have described the role of high-resolution mass spectrometers are required to univo- the ion ratio in the reconnaissance of pharmaceutical cally identify targeted compounds. Finally, libraries of compounds in aquatic environment using LC-MS. Estab- compounds, including MS-based information such as ac- lishing performance criteria for the global reconnaissance of curate mass, isotopic patterns, and collision-induced frag- pharmaceuticals is important since it minimizes the oc- mentation, are strongly demanded together with studies currence of false-positive and false-negative detection. Based regarding the establishment of recognized analytical per- on these assumptions, a performance criterion was disclosed formance criteria to assess the occurrence of residues in the by the authors and applied to several equal-to-real samples. environment. Mass spectrometric imaging is another For environmental assessment, in situ radiogenic isotope emerging powerful analytical technique that can be applied determinations with microscale resolution can represent to perform analyses of multiple molecules in complex a powerful tool especially for geological and life sciences: S. samples without labeling, thus providing a distinct advan- Di Salvo et al. have presented a detailed methodological tage over preexisting methods for label-free and simulta- description of the analytical procedure from sampling to neous detection of drugs and metabolites. elemental purification and Sr-isotope measurements. $e 2 Journal of Analytical Methods in Chemistry proposed method offers the potential to attain isotope data at the microscale on a wide range of solid materials with the use of minimally invasive sampling. Mass spectrometry plays a pivotal role also in the forensic field: in this context, N. Mogollon and coworkers have reviewed the recent developments in MS for forensic analysis focusing their attention to the identification and quantification of drugs of abuse in biological fluids, tissues, and synthetic samples. Both the most common methodologies and the new methodologies used for screening and target forensic analyses are reviewed, thus including high- resolution MS as well as the use of ambient ionization ion sources for high throughput and real-time monitoring. When very complex matrices have to be analyzed, multidimensional chromatography coupled to mass spectrometry can offer increased selectivity and separation power to solve different analytical problems: Y. G. Ahn et al. compared the performances of gas chromatography with quadrupole mass spectrometry and GC × GC-TOFMS for quantitative analysis of eighteen target polycyclic aromatic hydrocarbons in ambient aerosol. Although similar results were obtained in terms of both detection and quantitation limits, a larger number of analytes were identified by using the GC × GC-TOFMS method, thus suggesting that comprehensive two-dimensional gas-chromatography coupled to mass spectrometry such as GC × GC-TOFMS could be applicable to atmospheric and related sciences with simultaneous target and nontarget analyses in a single run. Mass spectrometry can be considered the technology of the future for medicine: its capabilities in biomarker discovery, development, and validation suggest the implementation of MS instruments in clinical labs. Nowadays, MS-based lab detection methods are increasingly used in hospital labs as well as in legal medicine: in this context, R. Hosli¨ and coworkers have compared the analytical performances of a GC-MS and a LC-MS/MS method, respectively, for the determination of phenytoin in different body compartments, i.e., blood, saliva, and human brain dialysate. $e LC-MS/MS method proved to be more sensitive than the GC-MS procedure, being also less time-consuming and requiring small amount of sample. Finally, L. Huang et al. have reported the role of trichloroacetic acid in enhancing the MS signal of tobramycin. Using a simple dilution with trichloroacetic acid as pairing reagent, a sensitive LC- MS/MS method was developed and validated in a bacterial medium.

Conflicts of Interest $e editors declare that they have no conﬂicts of interest.

Federica Bianchi Leopold Ilag Veronica Termopoli Lucia Mendez Hindawi Journal of Analytical Methods in Chemistry Volume 2018, Article ID 4142527, 17 pages https://doi.org/10.1155/2018/4142527

Review Article New Advances in Toxicological Forensic Analysis Using Mass Spectrometry Techniques

Noroska Gabriela Salazar Mogolloń ,1,2 Cristian Daniel Quiroz-Moreno,1 Paloma Santana Prata,2 Jose Rafael de Almeida,1 Amanda Sofıá Cevallos,1 Roldań Torres-Guie´rrez,1 and Fabio Augusto2

1Ikiam-Universidad Regional Amazo´nica, Km 7 Via Muyuna, Tena, Napo, Ecuador 2Institute of Chemistry, State University of Campinas, Cidade Universita´ria Zeferino Vaz, 13083-970 Campinas, SP, Brazil

Correspondence should be addressed to Noroska Gabriela Salazar Mogollo´n; [email protected]

Received 8 March 2018; Revised 30 May 2018; Accepted 12 July 2018; Published 29 August 2018

Academic Editor: Veronica Termopoli

Copyright © 2018 Noroska Gabriela Salazar Mogollo´n et al. .is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

.is article reviews mass spectrometry methods in forensic toxicology for the identification and quantification of drugs of abuse in biological fluids, tissues, and synthetic samples, focusing on the methodologies most commonly used; it also discusses new methodologies in screening and target forensic analyses, as well as the evolution of instrumentation in mass spectrometry.

1. Introduction monitoring (SRM). .is latter one is the most widely used because of its increase in the specificity, selectivity, and .e development of mass spectrometry methods has offered detectability; however, the analyses become too time- new possibilities for forensic toxicology analyses, where the consuming when a previous chromatographic separation identification and quantification of drugs of abuse are the and sample preparation are required [6, 7]. most concerning issues in the forensic science [1]. .e Moreover, ionization mass spectrometry techniques prevalence of drug addiction and abuse in the population such as direct analysis in real time (DART), desorption worldwide is significantly high, resulting in one of the main electrospray ionization (DESI), low-temperature plasma causes of high criminal activities [2]. .e excessive use of (LTP), desorption atmospheric-pressure photoionization psychotropic substances, natural drugs, hallucinogens, and (DAPPI), paper spray (PS), touch spray mass spectrometry most recently “new psychoactive substances,” which are (TS-MS), more recently in toxicological analysis laser diode designed from skeletons of some natural drugs previously thermal desorption (LDTD), and atmospheric solids analysis known, are the main focus of the development of new probe (ASAP) have gained popularity as they can be used analytical methodologies, where mass spectrometry has had with less or even without sample preparation [8–10]. a key role [3, 4]. When a toxicological analysis needs to Nevertheless, depending on the matrix sample, compounds identify and quantify metabolites from unknown drugs, with identical patterns of fragmentation cannot be identi- a screening can be performed by coupling different chro- fied; this is the reason why more development in mass matography techniques, such as liquid and gas chroma- spectrometry needs to be conducted in order to provide tography to mass spectrometry. In cases where an increment relevant information that can help solve a crime [11]. In this in the signal/noise ratio (S/N) is necessary, and the structure sense, this review presents the main current applications of of the compound is known (target analysis) [5, 6] an ad- mass spectrometry for the control of drugs of abuse and the ditional selectivity can be provided using tandem mass discovery of synthetic drugs in biological and synthetic spectrometry (MS/MS) in ion products or selected reaction matrices; besides, methodological limitations as well as 2 Journal of Analytical Methods in Chemistry innovative methodologies to enhance forensic toxicology amphetamines, and ketamines in human hair. .ese analyses analysis are discussed, examining the current literature in were capable of providing more sensitivity at low concen- the past eight years. tration of pictogram (pg) using only 25 mg from the sample and improving the detection of compounds during the 2. Chromatography and Mass Spectrometry analysis owing to the electronegative moieties. .e strategy also avoided wrong results and misinterpretations obtaining 2.1. Conventional MS Methods. .e coupling of chroma- lower limits of detection, in comparison with the use of only tography techniques with mass spectrometry has been traditional GC/EI-MS in mode SIM; therefore, NCI can serve widely used in drugs of abuse analysis, especially when the as a complementary technique in order to improve the screening of the sample is needed, having separation sensitivity during the analysis [14]. techniques such as gas chromatography-mass spectrometry .e use of a miniaturized analytical method is the aim (GC-MS), liquid chromatography-mass spectrometry (LC- during the development of new analytical methods, and the MS), liquid chromatography-mass spectrometry in tandem analysis of drugs of abuse by mass spectrometry is not the (LC-MS/MS) and, more recently, two-dimensional gas exception. Most recently, GC-MS methodologies have used chromatography-MS (GC × GC-MS) as the most commonly cold EI to analysis of heroin and cocaine [15, 16]. Here, the used. GC has an interface known as a supersonic molecular beam Normally, the analyses of nonobjective analytes, after the (SMB) where the ionization vibration cold sample is in an chromatographic separation, have the same steps to follow. axial fly-through ion source configuration (Figure 1(a)), First, a scan is performed by the mass spectrometer in order providing mass spectra with enhanced molecular ions that are to identify or recognize some compounds of interest; then, it compatible with reference libraries, and the range of com- is necessary to perform a selected ion monitoring (SIM) pounds are amenable to GC-MS compounds. Additionally, [12, 13] in order to increase the sensitivity and selectivity of this configuration allowed the increment of the flow rate in the analysis, in which only fragments of a specific group of GC-MS without declines in the sensitivity in the analysis in molecules are monitored, resulting in an increased S/N. the EI source, since the fly-through ion source sensitivity is Consequently, this technique is the most widely used in fully independent on the column flow rate; therefore, column quantitative analysis of compounds. flow rate increase is automatically offset by a corresponding .e focus in this section is on the most recent and in- reduction in the helium make-up gas flow rate, the supersonic novative analyses that have been performed using mass nozzle backing pressure, and the SMB flow rate are stabilized. spectrometry coupled to chromatographic techniques, in- .e authors considered this aspect during the determination cluding all the new methodologies developed in toxicological of heroin and cocaine in paper money and composite drug analyses. Table 1 provides a summary with their advantages powders using column flow programming as a tool to further and disadvantages, and Figure 1 shows these methodologies reduce the time of analysis. With this method, the time of as well. analysis decreased, allowing the use of a column flow from 1 mL/min to 32 mL/min and the use of relatively small column dimension (5 m 0.25 mm) [55]. 2.1.1. GC-MS. .e advances in techniques using MS coupled However, if the analysis aims to identify target com- to gas chromatography have not been very significant due to pounds, and if specific fragments of a molecule are known, it the type of analyte that can be analyzed using this chro- is possible to increase the S/N with the use of mass spec- matographic technique (low molecular weight, volatiles). trometry in tandem (MS/MS). GC-MS/MS is commonly Even though high molecular weight compounds can be used in SRM and product ion scan modes with collision- derivatized and analyzed by GC, the sample treatment is not induced dissociation (CID). On the one hand, an ion pre- appealing for the forensic toxicological analysis of drugs of cursor is generated into the collision cell during SRM mode, abuse where the quickness of analysis is fundamental. For and then one ion product is monitored—this monitoring is this reason, most of the advances using GC-MS focus on the also called transitions—this mode is widely used in quan- resolution and separation capacity during the analysis. titative analysis because of its selectivity. On the other hand, However, toxicological analysis methods in various matrices product ion scan consists of scanning product ions once the are well established and widely used in the analysis of drugs, molecules are fragmented in the collision cell, generating, as in order to confirm forensic toxicology from samples of a result, high reliability results due to the specificity of the blood, urine, saliva, and hair, among others, during specific monitored transitions. .is method is generally used for screening analysis, demonstrating high selectivity, de- transition optimization and the creation of libraries in tectability, and robustness. MS/MS. .us, these analyses can obtain an unequivocal In this sense, in order to improve the detection and identification of the eluted analyte. For example, this method identification of compounds using GC-MS, negative and identified methamphetamines in blood and urine with positive modes of analysis in MS have been integrated, taking a simple and quick LLE and derivatization, as well as advantage of the stability of the fragments after a positive or managed to differentiate between them [56]. negative ionization. For instance, Wu et al. used GC-MS with Versace et al. used GC-MS to perform a screening of electron impact ionization and negative chemical ionization unknown compounds without an excessive sample preparation (GC/NCI-MS) and the traditional GC-MS with electron in urine samples and GC-MS/MS with the purpose of in- ionization mass spectrometry (GC/EI-MS) to analyze opiates, creasing the specificity using SRM transitions, identifying Journal of Analytical Methods in Chemistry 3

Table 1: Main modifications and modes of analysis applied in mass spectrometry in forensic toxicological analysis. Ionization techniques in mass spectrometry coupled to separations techniques Type of MS analysis Advantages Disadvantages References (i) Better results are provided when (i) It provides more sensitivity at the technique is combined with EI-MS low concentration (pg) based on the in order to obtain more structural Negative chemical stability of electronegative moieties. information. [14] ionization (ii) Avoids wrong interpretations of (ii) .is method requires an additional correct results reducing time reagent for the ionization; methane consumption. is commonly used. (i) It can be considered as a miniaturized analytical method because of the interface that it uses and the supersonic molecular beams through analysis with short columns and high column flow rates. (ii) Can provide enhanced molecular ions to much larger and more polar compounds with GC, using the same library to EI-MS (NIST). Cold electrospray (i) Additional instrumentation is (iii) .e flow rate can be increased up [15, 16] ionization required. to 100 mL/min, and its fly-through ion source sensitivity is fully independent from the column flow rate. (iv) In this method, the nozzle flow rate is constant; as a result, the cold EI fly-through ion source is unaffected by the column flow rate, unlike any other ion source. (v) .e use of GC-MS with cold EI has no limitations for the column used. (i) .e ionization of solutes occurs upon the polarization of neutral, solvent molecules, which makes it a highly sensitive method. (ii) .e electrostatically charged surface increases the ESI ionization efficiency. (iii) When it is used with ESI, the efficiency of proton-transfer (i) SACI is used to maximize the ionization reactions is enhanced by sensitivity in the analysis of highly Surface-activated the polarization of neutral solvent polar compounds, but data about less [17] chemical ionization molecules or by charged solute polar compounds have not been molecules induced by the proximity of revealed until now. the charged surface. (iv) .e solvent and the analyte ions are better focused towards the analyzer. (v) .e increase in signal intensity provides an increase in sensitivity, because there is a reduction in the chemical noise observed in the mass analyzer. 4 Journal of Analytical Methods in Chemistry

Table 1: Continued. Ionization techniques in mass spectrometry coupled to separations techniques Type of MS analysis Advantages Disadvantages References New methods of analysis used in mass spectrometry (i) .is method monitors the analytes (i) It is necessary to maintain the only around the expected retention analyte analysis in the same polar time, decreasing the number of mode since a switch of polarity within concurrent MRM transitions, allowing a single run would reduce the both the cycle and the dwell time, sensitivity and accuracy of which can be optimized in order to quantification with the applied MS obtain higher sensitivity, accuracy, instrumentation. and reproducibility. (ii) .e retention time must be informed, optimized, and defined with (ii) dMRM allows the monitoring of reference standards using established more MRM transitions in a single run chromatographic conditions if it is Dynamic multiple reaction without compromising data quality. possible. If the retention time drifts, [17–20] monitoring mode of analysis this might result in an incomplete peak definition and quantitation. (iii) .e dwell time is intelligently optimized by association with the delta retention time. Additionally, information about delta retention time and retention time are key to (iii) It is necessary to optimize the MS maximize the dwell time and conditions for the all transitions. increasing sensitivity. (iv) .is method gives the possibility of applying simultaneous quantifications of multiple components. Ambient ionization techniques in mass spectrometry (i) During analysis of drugs in (i) Direct analysis with high-velocity biological matrix with a high amount nebulizing gas. of salt, the suppression ionization effect is elevated. (ii) .e ion source geometry affects the Desorption electrospray ionization- dynamic of the splashing mechanism [8, 21–23] mass spectrometry (ii) .e selectivity and sensitivity of resulting in changes in droplet size, this technique can be increased by charge, and analyte dissolution extent. a pretreatment sample. (iii) A high velocity of nebulization can mechanically ablate delicate samples/powders. (i) High suppression ionization can be found depending of the biological (i) Matrix with high salt content do Desorption atmospheric-pressure matrix. not provide an elevated suppression of [8, 21–25] photoionization (ii) Sample preparation is commonly ionization. needed in order to avoid suppression of ionization. (i) It is commonly used in the analysis of drugs of low molecular weight; (i) Compounds of high molecular therefore, its sensitivity depends on weight may need derivatization. analyte volatility. (ii) Its sensitivity depends of the (ii) .e geometrical configuration of temperature of the ionization region; Direct analysis in real time the ion source is simple and robust for [4, 8, 11, 26–35] therefore, the higher the temperature is its operation. the higher the risk of damage is. (iii) Its reproducibility depends on the (iii) Pretreatment of sample can position of the sample inside the ion increase the selectivity of the analysis source, which represents a big problem in complex biological samples. in the quantification of the analysis. Journal of Analytical Methods in Chemistry 5

Table 1: Continued. Ionization techniques in mass spectrometry coupled to separations techniques Type of MS analysis Advantages Disadvantages References (i) It is possible to perform direct analysis without sample preparation. (ii) .e instrumentation is simple, and its configuration provides low (i) .is technique is exclusively used consumption of discharge gas and the Low-temperature plasma with small organic molecules with low [8, 36] possibility of using air as the discharge to moderate polarity. gas. (iii) High sensitivity and sensitivity can be obtained without pretreatment of the samples. (i) It can be coupled to mass spectrometric imaging (MSI) in order to obtain the distribution spectra of the target. (ii) A mode of analysis called “dynamic pixel” can be used to obtain an imaging method that is faster to do a screening of the compounds. (iii) .e analysis does not need sample preparation. .is method is based on Matrix-assisted laser desorption (i) Quantitative analysis has not been a direct analysis over the sample. [8, 37–41] electrospray ionization carried out until this present date. (iv) .e sensitivity of the analysis can be improved using a specific matrix. For example, umbelliferone matrix obtained better results in the analysis of methamphetamine in hair than the common matrices CHCA or DHB. (v) .e technique has been tested along with MAMS, and it is possible to cause reproducibility of the signal with this technology. (i) It can be coupled to mass spectrometric imaging (MSI) in order to obtain the distribution spectra of the target. Metal-assisted secondary ion mass (ii) .e limits of detection are lower (i) Quantitative analysis has not been [8, 42, 43] spectrometry than those obtained with MALDESI carried out. and also compared with the ones with LC-MS/MS. (iii) It is not necessary to perform preparation of the sample. (i) .is technique can analyze a wide (i) It has a high matrix effect on most of range of molecules, from small to large the drugs. biomolecules. Paper spray [8, 44–51] (ii) .e use of a pretreatment of the (ii) .e paper can extract impurities sample can enhance the sensitivity of from the surface and cause the the analysis. suppression of ionization. (i) Methods of introduction of samples High-performance ion mobility such as a chromatographic separation (i) Direct analysis can result in [50, 51] spectrometry can be used to minimize the suppression of ionization. suppression of ionization. 6 Journal of Analytical Methods in Chemistry

Table 1: Continued. Ionization techniques in mass spectrometry coupled to separations techniques Type of MS analysis Advantages Disadvantages References (i) Separation conditions of the target analysis can be selectively transmitted (i) .is technique is rarely being into a mass spectrometer. implemented in commercial devices, Differential mobility spectrometry- (ii) It can be considered as an and it is not known yet whether it can [52] mass spectrometry separation ionization technique coupled to be used to establish profiles of drug a separation method that has a small mixtures in complex biological interface which gives results in few samples. seconds. (i) .e substrate (medical swabs) used can serve as a sample collection tool; thus, ionization helps in the analysis of (i) .e drying step of this substrate Touch spray solid or liquid samples without represents the most time-consuming [9] pretreatment. part of the analytical protocol. (ii) .e TS-MS can allow noninvasive and direct sampling of neat oral fluids. (i) It is not possible to perform a simple interchange between negative and positive modes of ionization. (ii) Effects of interferences in complex (i) .e method is completely Laser diode thermal desorption biological samples must be explored, [53] automatic. and more sample preparation is necessary before the liquid samples are transferred towards the capillary surface. (i) Effects of interferences in complex (i) It is possible to perform analysis of biological samples must be explored solids and liquids easily. with more detail. Atmospheric solids probe analysis (ii) .is technique provides better [53, 54] (ii) .is design allows the possibility of sensitivity during the analysis of small- positive/negative switch during the molecule drugs, decreasing the analysis analysis. of the high-molecule compounds.

He make-up gas MS

Electrospray ion source GC Nebulizing gas

Heated nitrogen drying gas Supersonic agent Cluster CI agent molecular beam in molecular Calibration Solvent spray Charged droplets SMB interface lenses Ion beam Dielectric capillary entrance Surface metallic Ion mirror and Heated nitrogen HSI surface Filament Repeller Ion cage Transfer line with supersonic nozzle Skimmer SMB vacuum chamber (a) (b)

Figure 1: Main advances in mass spectrometry coupled to chromatographic technique in toxicological analysis are (a) gas chromatography interface-supersonic molecular beam with ionization vibration cold sample to mass spectrometry and (b) electrospray ionization and surface-activated chemical ionization to mass analyzed coupled with liquid chromatography.

54 drugs (i.e., 11-nor-9-carboxy-Δ9-tetrahydrocannabinol, ion trap/tandem mass spectrometry. Here, CID was cocaine, hydrocodone, and ﬂurazepam) [57], while used to adjust the breakage of the cannabinoid frag- Emidio et al. developed a new methodology to determine ment (ion precursor) and improve the detectability of the cannabinoids in hair using 10 mg of sample and head- technique, demonstrating an excellent linearity range between space solid-phase microextraction (HS-SPME) and GC 0.1 and 8.0 ng/mg with a limit of quantiﬁcation (LOQ) of Journal of Analytical Methods in Chemistry 7

0.007–0.031 ng/mg and 0.012–0.062 ng/mg, which are smaller signals were optimized using a source block temperature of than the cutoff value established by the Society for Toxico- 500°C and an on-spray in the determination of Δ9-tetra- logical and Forensic Chemistry (GTFCh) [58]. hydrocannabinol, cannabidiol, and cannabinol in 50 mg of On the other hand, ethyl glucuronides (a biomarker of 179 hair samples. .is method allowed the identification of ethanol), commonly used in the detection of chronic and one new synthetic cannabinoid, obtaining an LOQ of around excessive alcohol consumption, were identified using 0.07 pg/mg and 18 pg/mg in the analysis [63]. MS/MS operating in NCI-MS and SRM mode, obtaining High-resolution (HR) MS has been used successfully differences between teetotalers and moderate drinkers, along with LC in the drug of abuse analysis. .e resolving according to the current cutoff (i.e., 7 pg/mg hair). In this power and the high mass accuracy obtained with HRMS case, the use of negative mode provided an enhanced sen- were advantageous in the analysis of complex matrices and sitivity in low concentration samples which were combined data acquisition in a targeted and nontargeted manner in with the specificity of the fragments in the SRM analysis. order to decrease the number of interferences caused by .erefore, a better analytical selectivity and S/N were biological matrix in the drugs analysis. In this case, the achieved along with long-term markers for the detection of authors used UHPLC-HR-TOFMS to analyze cannabinoids chronic and excessive alcohol consumption [6]. and cathinones in 1 mL of urine. During the analysis, In the same manner, GC-MS/MS has been used to dif- a broad-band collision-induced dissociation (bbCID) was ferentiate among important isomers such as methox- used with the purpose of providing a confirmation-level yethylamphetamines and screening featuring both high sensitivity and wide scope. .e monomethoxydimethylamphetamines, as synthetic drugs precursor ions were fragmented in the collision cell without without derivatization. Using CID and SRM, the specificity of preselection, and the analysis allowed the identification of 75 the fragments obtained provided intensity differences in compounds with cannabinoids spectra database with cutoff product ions among the isomers, enabling mass spectrometric concentration values of 0.2–60 ng/mL and cathinones 0.7– differentiation of the isomers [59]. One the contrary, GC × GC 15 ng/mL, respectively [64, 65]. as a previous treatment can be used in order to increase .e methods of ionization are fundamental to ensure the sensitivity, detection, separation, and resolution. In this sense, correct ionization of the sample and, therefore, the identi- GC × GC-MS was used in the determination of cannabinoid- fication of the compounds, especially in complex matrices. like drugs in 1 mL of postmortem blood, which present Wang et al. performed an analysis of cocaine and their a challenge due to the matrix interferences in endogenous metabolites under three types of ionization such as elec- lipophilic compounds, proteins, drug degradation/formation, trospray ionization (ESI), atmospheric-pressure chemical and production of artefacts. With this technique, a limit of ionization (APCI), and atmospheric-pressure photoioniza- detection of 0.25 ng/mL for 11-hydroxy-Δ9-tetrahydrocan- tion (APPI) in order to evaluate the chemical suppression nabinol was obtained [60]. during the analysis of 17 illicit drugs in 100 µL of oral fluids .e same methodology was applied in oral fluid samples, using UHPLC-MS/MS in mode SRM. .e authors found where it is common to have a small volume of samples, and the that ESI presents the smallest ion suppression for all cocaine concentration of some drugs is usually lower, which can metabolites analyzed facing the APPI and APCI mode. complicate the analysis [61]. .e compounds were identified However, the method developed obtained LOQs in ESI, using GC × GC-MS with cold trapping and NCI-MS, obtaining APCI, and APPI in a range from 0.11 to 1.9 ng/mL, 0.02 to a limit of detection of 0.5 ng/mL [61]. Additionally, GC × GC 2.2 ng/mL, and 0.02 to 2.1 ng/mL, respectively. .e authors coupled with time-of-flight-MS (GC × GC-TOFMS) was used recommended further investigation to determine the causes to analyze codeine, morphine, and amphetamines in of higher ion suppression in APCI and APPI on ESI in oral sample extracts from hair suspected of containing various fluids, since ESI may suffer important matrix effects, as it is drug compounds. .e analytical technique also identified widely known. Nevertheless, they state that APCI and APPI metabolites such as cocaine, diazepam, and methaqualone, probes evaporate inlet solutions and ionize analytes via gas- which are not included in the target analysis [62]. phase chemistry and, consequently, are less affected than ESI. For example, oral fluids may contain many salts and small molecules partitioned from plasma instead of mac- 2.1.2. LC. .e use LC as a versatile separation technique romolecules, which can lead to an increase in ion sup- (volatile and nonvolatile analytes) has improved the de- pression in APCI and APPI. As a result, authors suggest the tection and quantification of analytes such as amphetamines, use of ESI in this type of analysis [66]. benzodiazepines, hallucinogens, cannabinoids, opiates, co- Liquid chromatography has also taken advantage the caine, designer drugs, pharmaceutical products, or illicit benefits of negative mode in mass spectrometry. In this case, drugs in several matrices. LC coupled to (HR)-MS was used along with Orbitrap Ultrahigh-performance liquid chromatography has been technology in the analysis of metabolites of drugs such as used along with tandem mass spectrometry (UHPLC- cocaine, ephedrine, and morphine in urine. .e analyses were MS/MS) operating in SRM mode in order to establish an performed in full-scan mode with positive/negative switching, individual ion transition ratio to each analyte. .us, each and subsequently making use of a selective screening through analyte is infused into the electrospray capillary, and the data dependent acquisition (DDA) mode, resulting in a fast declustering potential was adjusted to maximize the in- analysis. Additionally, the risk of false-negative results caused tensity of the protonated molecular species [M + H]+. .e by ion suppression or isomer overlapping could be reduced by 8 Journal of Analytical Methods in Chemistry including metabolites and artefacts, as well as recording in the directly on the sample. .e main difference between positive and negative modes [67]. chromatography-MS methods is the sample introduction. Recently, new methods of analyses in MS/MS have been MS instrumentation is assembled by ion sources, mass coupled with LC techniques. Dynamic multiple reaction analyzer, and detector [13]. However, the challenge in fo- monitoring (dMRM) has been used in toxicological analysis, rensic analysis is the possibility of decreasing the time and and it is recognized by the use of a timetable based on the cost of analysis per sample. A primary analytical focus in retention time for each analyte. Such technique monitors the toxicology is determining the presence or absence of drug analytes only around the expected retention time, and decreases metabolites in biological samples. In this sense, the use of the number of concurrent SRM transitions, also known as ambient ionization technique mass spectrometry has multiple reaction monitoring (MRM), allowing both the cycle allowed the analysis of the entire sample without an ex- time and the dwell time to be optimized to the highest sensitivity, cessive preparation of sample. .ese techniques make accuracy, and reproducibility [18]. For example, a quantitative possible the concept of open-air surface analysis directly LC-MS/MS method has been developed for the simultaneous under ambient conditions, being particularly useful for determination of 17 antipsychotic drugs in human postmortem surface analysis of solids, avoiding many, if not all, sample brain tissue; these drugs are of forensic interest because they preparation steps typically required [8]. have been associated with sudden death cases. HRMS is widely employed in the coupling with ambient In this method, the analysis was performed operating on mass spectrometry currently because of its capability of dMRM mode, using ESI+. Calibration curves prepared in measuring accurate masses and differentiating among the spiked brain tissue were linear in the range 20–8000 ng/g compounds with identical nominal masses, providing (R2 > 0.993) for all drugs, except olanzapine [19]. Besides, a comprehensive full-scan MS and MS/MS for the search for LC-MS/MS in dMRM mode was used by Shah et al. in order any analyte without sample pretreatment. .is provides to identify around 200 drugs/metabolites, such as meth- accurate m/z values that can be used to generate chemical amphetamine, amphetamines, ephedrine, and cocaine in formulas with high mass accuracy (<5 ppm mass error) [70]. hair samples [20]. .is method proved an interesting al- .erefore, HRMS can be theoretically applied in different ternative for fast analysis of drugs. All these analyses were configurations with interchangeable ionization sources and performed in one chromatographic run (i.e., 8 min), sophisticated data acquisition capabilities, making HRMS showing a high sensitivity and accuracy [20]. one of the preferred techniques for the analysis of new drugs With the purpose of identifying cannabinoids, such as [71]. Δ9-tetrahydrocannabinol, Conti et al. coupled LC along with Ambient ionization mass spectrometry can be divided two ionization systems, electrospray ionization, and surface- depending on the desorption technique, which will be activated chemical ionization (ESI-SACI-MS) to several discussed in the next section about the most used procedures types of mass analyzer (ion trap, triple quadrupole, and in the analysis of drugs of abuse. Table 1 and Figure 2 present Orbitrap) to improve the detection of 11-nor-9-carboxy- the advantages and disadvantages of these techniques or tetrahydrocannabinol in biological samples (urine and hair) procedures. operating in SRM mode. .is coupling consists in a metallic surface that keeps a fixed voltage and that is inserted into a commercial ESI source (Figure 1(b)). .is electrostatically 3.1.1. Ambient Ionization Technique Mass Spectrometry: charged surface is able to improve the ESI ionization effi- Desorption by Solid-Liquid Extraction. In these techniques, ciency, and it increases the ion focusing efficiency towards the desorption occurs by solid-liquid extraction followed by the mass spectrometric analyzer. .e authors show that the ESI-like ion-production mechanisms; the ionization can be sensitivity provided was better with SACI-ESI than with the performed by DESI (Figure 2(a)) or DAPPI (Figure 2(b)) [8]. classical ESI approach alone [17, 68]. DESI-MS is used in forensics due to its ability for in situ Furthermore, an ultrafast and sensitive microflow liquid analysis. However, direct analyses that involve DESI and chromatography-MS/(MFLC-MS/MS) was used to quantify DAPPI are less common in toxicological analysis due to the hallucinogens such as LSD and their metabolites in 500 µL of interferences which can be caused by the suppression of plasma in order to miniaturize and accelerate the analysis of ionization product of the matrix effects in the samples; drugs of abuse using LC techniques; this coupling is known by its therefore, adding an additional step in the preparation of the decrease in run numbers, a higher ionization yield, and reduced sample is often required. For instance, matrix-suppression ion suppression/enhancement effects. Here, the MS ion trap effects were studied within direct analysis of benzodiaze- operated in product ion scan and SRM mode in order to perform pines and opioids from 1 mg/mL of urine with DESI-MS and the quantification along with a dynamic fill-time trap; this DAPPI-MS [21]. .e authors found that the urine matrix method allowed sensitive detection and fast analysis, obtaining affects the ionization mechanism of the opioids in DAPPI- LOQs corresponding to 0.01 ng/mL for all analytes [69]. MS and favors the proton transfer over charge exchange reaction. However, the sensitivity of the drugs in the solvent 3. Current Analytical Approaches to matrix was at the same level in DESI-MS and DAPPI-MS Target Analyses with limits of detection of 0.05–6 µg/mL, along with a decrease in sensitivity for the urine matrix that was higher with 3.1. Mass Spectrometry. Mass spectrometry is the preferred DESI (typically 20–160-fold) than with DAPPI (typically technique when the aim is to perform a quick analysis 2–15-fold), indicating better matrix tolerance in DAPPI over Journal of Analytical Methods in Chemistry 9

Nebulizing gas Nebulizing gas Solvent (N2) Solvent (N2) UV lamp Atmospheric pressure Mass spectrometry photoionization source Mass spectrometry

Mass analyzer Mass analyzer Charged droplets Electrical potential Electrical potential Charged droplets MS inlet (a) (b)

Analyte Vacuum Discharge gas interface

N2/He gas Metastable Electrode Dielectric Atmospheric inlet of MS He/N 2 barrier DesorbedD ions HV Ion/neutral Heating Surfaceface Discharged chamber separation Analyte

Paper Mass analyzer Electrical potential MS inlet Sample HV

Plume of the matrix and sample

Sample and matrix (e) (f) Differential Mass spectrometer mobility cell Sample Drift field Transporting gas Quadrupole + Ion source Aperture grid carrier gas Drift rings Drift gas in Ions Ion funnel Einzel Multiplier lens Ion gate Detector Ion source Drift region Detection (g) (h)

Desolvation or APCI source HV ASAP probe

Analyte Gas auxiliary + Solvent Hot N Glass capillary 2 tube MS inlet Sample Corona discharge pin MS inlet

(i) (j) Figure 2: Continued. 10 Journal of Analytical Methods in Chemistry

MS inlet IR laser beam Transfer tube transporting the neutrally desorbed analytes Corona needle discharge region

LazWell plate sample (k)

Figure 2: Main ambient ionization techniques used in toxicological forensic analysis. (a) DESI-MS, (b) DAPPI-MS, (c) DART-MS, (d) LTP- MS, (e) MALDI-MS, (f) PS-MS, (g) HPIMS, (h) IMS-MS, (i) TS, (j) ASAP-MS, and (k) LDTD-MS.

DESI. .is illustrates that urine contains high concentra- and liquid states, handling polar and nonpolar analytes with tions of salts in DAPPI, and the salts in the urine samples are masses below 1 kDa [26]. not efficiently evaporated from the sampling surface which Compounds are identified by combining information do not significantly interfere with the ionization [21]. about elemental compositions from exact masses and iso- DESI-MS also allowed the analysis of common drugs in topic abundances with fragment-ion mass spectra obtained urine samples such as pethidine, diphenhydramine, nor- by collisional activation. In toxicological forensic analysis, triptyline, and methadone using pretreatment sample by DART has been widely used in the detection of small drugs, liquid-phase microextraction (LPME). .ese selective ex- but its quantitation remains a big problem due to its minor traction capabilities of three-phase LPME provided a sig- reproducibility, which depends on the position of the sample nificant reduction in the matrix effects observed in direct inside the ion source, making the number of drugs that can aqueous LPME extracts [22]. However, some drugs such as be quantified very limited [27]. However, when it is nec- Δ9-tetrahydrocannabinol and cannabidiol that have iden- essary to obtain more details in the identification matrix tical fragmentation spectra presented significant in- based on the natural products, high temperatures of ionizing terference, resulting in the impossibility of an unequivocal gas can be used during the analysis, benefitting the reso- identification of each other [22]. Additionally, DESI-MS/MS lution of more complex spectra [28]. has been used coupled with solid-phase extraction (SPE) in DART achieved the detection of c-hydroxy butyrate order to analyze clenbuterol in urine specimens to detect without any sample preparation or other illicit synthetic doping; the authors mentioned that the suppression effects cannabinoid products coupling mass spectrometry DART- were minimized by SPE using DESI-MS/MS [23]. MS with CID analysis. .e use of fragments obtained by CID Moreover, DAPPI-MS has been used coupled with provided a sensitive and specific detection, increasing the quadurople-ion trap MS and MS/MS mode to analyze di- limit of detection to identify individual components and rectly herbal products such as Catha edulis, Phycybe showing the ions related to each synthetic cannabinoid, since mushoms, opium, designer drugs in tablets, confiscated drug the [M + H]+ precursor ions were still present in the mass samples of several forms as tablets, blotter paper, plant resin, spectra [29]. .us, an unambiguous differentiation of each and powder forms that contain meta-chlorophenylpiperazine, 3- species could be accomplished. [M + H]+ precursor ions fluoromethamphetamine, methylenedioxypyrovalerone, am- could also be used as a complement in the analysis of drugs phetamines, phenazepam, buprenorphine, and methylone. through screening in order to identify new and unknown DAPPI-MS proved a specific analysis without sample prepa- drugs. ration [24], showing that it is sufficient in most criminal cases DART-TOFMS detected alprazolam, which is one of where the main purpose is to do a qualitative screening [25]. the ingredients of the “Houston Cocktail,” containing hydrocodone/acetaminophen, and achieved an analysis with high mass accuracy [30]. Habala et al. identified six 3.1.2. :ermal or Chemical Sputtering Neutral Desorption. synthetic cannabinoids in methanolic extracts from solid .ese techniques involve metastable and reactive ions, in herbal material using a DART source coupled with a hy- which the species react with the analyte directly or indirectly brid ion trap—LTQ ORBITRAP—mass analyzer, discov- through proton- and charge-transfer reactions [8]. In the ering that the leaves have a greater concentration than the analysis of drugs of abuse, DART (Figure 2(c)) and LTP stems of the plant material [11]. (Figure 2(d)) can be used; the former is the most preferred in In the identification of new psychoactive substances toxicology forensic analysis and uses a negatively biased (NPS), DART-MS has had an important role. Gwak and point-to-plane atmospheric-pressure glow discharge at Almirall [4] performed a screening of 35 NPS in urine using lower currents, physically separated from the ionization DART coupled with hybrid TOFMS and ion mobility region by one or several electrodes. .e metastable species spectrometry (IMS), identifying synthetic cathinones with are formed within the discharge supporting gas that typically a single phenethylamine as the most common compounds. is He or N2, generating protonated water clusters. .e main .e analytes detected had an error within ±5 ppm, but advantage of DART is the analysis of samples in solid, gas, isomeric compounds could not be differentiated. Similarly, Journal of Analytical Methods in Chemistry 11

DART-TOFMS was used in order to detect synthetic can- time-of-flight (QTOF). .e authors found that only the nabinoid in botanical matrices like Coriandrum sativum, quadrupole-Orbitrap in high-resolution mode achievement Ocimum basilicum, and Mentha spicata [28]. In this re- distinguished THC in hair samples from endogenous isobaric search, intensive sample preparation was not required, just interferences [35]. .ose are important data since when the methanol dissolution, which is a method that allowed the resolution in the mass analyzers is low, the risk of obtaining identification of the synthetic cannabinoids such as AM-251 false/positive is high. and JWH-015. Although botanical samples exhibit relatively Different from DART, LTP was developed for direct complex mass spectral profiles, this did not prohibit the sampling ionization in chemical analysis using mass spec- identification of the target compounds. Additionally, the trometry. .e plasma here is generated by dielectric barrier DART-TOFMS analyses were conducted with different ionizing discharge (DBD) and a discharge of gas at low flow rate gas (helium) temperatures in order to determine the optimum (<500 mL/min), and a high-voltage to sustain the plasma in desorption temperature, and it was observed that higher an ambient environment [8]. .is technique has proved temperatures had the additional benefit of yielding more a powerful tool in direct analyses, exclusively with small complex spectra that could permit a more detailed identification organic molecules with low to moderate polarity. For this of the plant matrix based on the natural products [28]. reason, it is not commonly used in the analysis of illicit high Grange and Sovocool [31] developed a methodology for molecular weight drugs as it limits the analysis of unknown the extraction and clean-up of drugs in smoke deposited on drugs. However, LTP proved effective in the analysis of household surfaces so as to determine the exposure of the stomach fluid content of a diseased dog suspected to have patients to drugs of abuse using DART-TOFMS. A field died from ingestion of insecticide. Direct sampling ioniza- sample carrier and an auto sampler were used to minimize tion was applied in MS analysis and protonated Terbufos, the time per analysis. .e sampling was performed just with and Terbufos sulfoxide were observed [72]. .ese two cotton swab wipes with isopropanol, finding a quantification compounds are common in Terbufos-based insecticides, of each drug of around 0.025 µg/100 cm2. However, Δ9- which were suspected to be the cause of the death of the dog. tetrahydrocannabinol and nicotine had m/z 315 and m/z 163 Furthermore, the analysis of drugs of abuse in urine interferences, respectively. .e authors found that this in- and 25 mg of hair extract samples were systematically interference could be a sugar unit from the cellulose of the vestigated, where several drugs such as amphetamine, cotton-swabs. In spite of this interference, the method is benzoylecgonine, caffeine, cannabidiol, cocaine, codeine, highly recommended for the analysis of residues in clan- diazepam, ephedrine hydrochloride, heroin, ketamine, destine drug laboratories [31]. methadone, methamphetamine, morphine, and Δ9-tet- Moreover, phenethylamine, a synthetic drug known by rahydrocannabinol were identified obtaining a limit of its effects similar to LSD and its sublingual consumption via detection of around 10 ng/mL without any sample preparation blotter paper, was analyzed directly in the sample by DART- [36]. TOFMS. .is was studied in blotter paper street samples, and the results can be used in preliminary identifications, since this technique is extremely fast and advantageous for 3.1.3. Laser Desorption/Ablation. In these techniques, the the quick screening of unknown street samples in crime analytes are desorbed or ablated from a surface by an IR or laboratories [32]. Poklis et al. used DART coupled with UV laser with or without a matrix (Figure 2(e)). .e sample HRMS in the analysis of legal purchases on the Internet is subsequently merged with an electrospray droplet cloud or under the name “Raving Dragon Novelty Bath Salts and a plasma stream, depending on the ionization source used Raving Dragon Voodoo Dust” and found out that they [8]. When the source excites an exogenous matrix that contain methylone and pentedrone, respectively, which can cocrystallizes and has energy absorbent capabilities, it can be identified as unsupervised drug market [33]. coat the sample surface to be analyzed. .en, a laser adds More recently, DART has been developed using SPME- excess energy to the matrix-sample complex, where the fiber format for coupling nanogold surfaces with mass matrix absorbs laser energy to pass it to the sample and, spectrometry in order to perform an effective drug capture finally, to produce ions from analytes; this technique is called in toxicological matrices like methamphetamine, diazepan, matrix-assisted laser desorption electrospray ionization and alprozolam in human plasma. .e authors used LC- (MALDESI) [73]. In contrast to MALDESI, metal-assisted MS/MS and DART-MS/MS, in this case, coupling antibodies secondary ion mass spectrometry (MetA-SIMS) procedure to nanogold-coated wires. An antibody with cross reactivity adds small amounts of metals onto sample surface to en- to multiple drugs was used for simultaneous extraction of hance mass spectra analysis [42]. .ese two methods may a mixture of drugs. .e immunoaffinity nanogold is known provide an image coupled with ionization mass spectro- by its possibility of eliminating chemical noise. .e limits of metric imaging (MSI), which is a powerful technique to detection achieved with DART-MS/MS were comparable to obtain spatial information (distribution) of compound mass those observed with LC-MS/MS [34]. spectra. Different mass analyzers have been used to evaluate Porta et al. used MALDESI coupled to MSI in order to sensitivity and selectivity in the detection of Δ9-tetrahy- monitor the distribution of cocaine and its metabolites in drocannabinol (THC) from intact hair samples using DART 12 mL of extracts of intact single hair samples from chronic [35]. .e mass analyzers evaluated were an Orbitrap, users. .e acquisitions were performed applying rastering a quadrupole-Orbitrap, a triple quadrupole, and a quadrupole mode in the SRM mode on a MALDI triple quadrupole 12 Journal of Analytical Methods in Chemistry linear-fitted ion trap. .e time of analysis of an intact single .is method enhances the detection and sensitivity of target hair sample of 6 cm was of 6 min approximately. Cocaine drugs embedded in a hair matrix, achieving a detection level and its metabolites were identified and quantified, and the down to nanogram per milligram; for this reason, the au- results were obtained with a limit of detection of 5 ng/mL, thors compared the results with the obtained by LC-MS/MS, becoming an excellent methodology to detect cocaine but in this case with less sample amount required [40]. consumption [37]. In the same manner, matrix-assisted laser More recently, Kernalleguen´ et al. [41] have made desorption/ionization-mass spectrometry imaging (MALDI- possible the semiquantification of cocaine and its metabo- IMS) was used to rapidly screen longitudinally sectioned drug lites (benzoylecgonine, cocaethylene, and ecgonine methyl user hair samples for cocaine and its metabolites; using ester) in hair, using microarrays for MS and MALDI- continuous raster imaging, the optimization of the spatial MS/MS. So far, it is well known that the inhomogeneous resolution and raster speed were performed on cocaine- MALDI matrix crystallization and laser shot-to-shoot var- contaminated intact hair samples. Besides, the MALDI- iability make the quantitation more difficult; therefore, the MS/MS images showed the distribution of the most authors used a high-throughput MALDI method, along with abundant cocaine, using product ion scan as a mode of ana- an innovative high-density microarray for mass spectrom- lyzing. With this method, it is possible to obtain mass spectra etry (MAMS) technology. .is technology consists of with the main fragment of the molecule target. a sample preparation slide containing lanes of hydrophilic An SRM experiment was also performed using the spots, and an automated slider which drags a sample droplet “dynamic pixel” imaging method to screen for cocaine and over several small spots, with the purpose of achieving a range of its metabolites, in order to differentiate between homogeneous crystallization of the matrix-analyte mixture contaminated hairs and drug users. .erefore, these and, therefore, to a reproducible signal. In this manner, it methods are important when the imaging information on was possible to establish a calendar of consumption in only drug distribution is necessary, for example, in human hair 1 mg of hair with a great correlation, becoming an excellent without extensive sample preparation, or when labelling methodology when urgent results are required [41]. techniques are required. However, it should be noted that it However, metal MetA-SIMS was used to determine the only provides qualitative data about administered drugs, differentiation between systemic exposure and external through a pixelated representation [38]. MALDESI has also contamination that remains in the hair because of exposure been coupled with HRMS during the identification of 74 to drugs after following the protocols of decontamination drug samples which were detected using the ionic liquid (hair wash) [43]. .e authors reached a comparison of the matrix N,N diisopropylethylammonium α-cyanohydrox- results among MetA-SIMS, MALDESI-MS, and LC-MS/MS, ycinnamate. .is method allowed the identification of new showing that there is still cocaine detected after the washes of designer drugs, which come from the use of safrole as decontamination, using MetA-SIMS. MALDESI was in turn a precursor for the synthesis. Nevertheless, the result ob- inefficient for forensic hair analysis since no cocaine was de- tained presents weaker resolution and lower sensitivities, tected after decontaminating the samples. LC-MS/MS detected leading to lower peak intensities. .e authors affirm that this 5 ng/10 mg in the sample after the washing. Finally, the authors limitation is a consequence of the matrix, since this can be concluded that the washing protocols are not reliable, because related to the formation of adducts, but the matrix may be external cocaine can migrate into the hair, and recommended enhanced by adding specific cations and anions. Further a simple analysis of images which makes the evaluation of the investigations to improve ionization through matrix addi- differences among hair samples contaminated externally and tives are still necessary in the field. Another limitation of this the interpretation of the correct results easier [43]. methodology is the impossibility to distinguish drug position isomers, such as methamphetamine and 4-methyl- amphetamine, as well as structural elucidation of unknown 3.1.4. Other Methods of Ionization. Paper spray (PS) tech- compounds. .e authors recommend the combination nique was introduced in 2009 and has been used in the between this methodology and bioinformatics software tools development of a wide range of quantitative and qualitative which provide untargeted compound searches, even if re- applications. Here, the sample is deposited in the paper with spective HRMS spectra are not included in a library just a sharp point, and ions are produced by voltage applied, based on the precursor ion fingerprinting [39]. while the substrate is held by a metal clip in the paper and On the contrary, MALDI-MSI and MALDI-Fourier placed in the front of the inlet of a mass spectrometer. .en, transform ion cyclotron resonance (FTICR-MS) have also the front mass spectrometer performs the detection after the been used for mapping and direct detection of metham- sample elution, which can be carried out in the same manner phetamine in longitudinal sections of the single hair sample of paper chromatography, but with a direct sample injection in positive mode, in which umbelliferone was used as to the mass spectrometer (Figure 2(f)). In this technique, a matrix. .is matrix has the advantage of being hydro- a wide range of chemicals can be ionized by paper spray, phobic and capable of assisting in the ionization of meth- from small molecules to large biomolecules [8]. amphetamine in hair. .e authors observed that the Paper spray ionization coupled to high-resolution tandem detection and sensitivity provided by this matrix is higher mass spectrometry (PSI-HR-MS/MS) have also been used in than α-cyano-4-hydroxycinnamic acid (CHCA) or 2,5- order to validate a screening of drugs in urine such as codeine-6- dihydroxybenzoic acid (DHB). In addition, the distribution glucuronide, diclofenac, among others, and to validate a com- semi-quantitative of methamphetamine can be performed. prehensive urine screening. Nevertheless, the procedure showed Journal of Analytical Methods in Chemistry 13 high matrix effects for most drugs, but also acceptable limits of interfering with matrix components were observed, so the identification that have the potential of reducing workload. authors recommended considering some previous steps However, the authors recommend its implementation as before sample preparation. For instance, the authors in- a promising alternative to conventional procedures, but they troduced a sample into a drift tube via pulse Bradbury– warned that there is a risk of false positive/negative results Neilson ion gate and operated it in positive mode, and the caused by mixed spectra during the detection of low concen- ions passed to desolvation to be separated [50]. .is method trations. .erefore, some problems should be solved before achieved a resolving power double than the currently ac- implementing it in routine analysis [44]. cepted method without an excessive necessity of sample .in-layer chromatography (TLC) has been also used as preparation [51]. an introductory sampling method combined with PS-MS to Ion mobility-based separation methods can be combined analyze cocaine and its adulterants in 10 µL of sample. .is with mass spectrometry (IMS-MS) in order to minimize analysis obtained promising results in which the limit of chemical suppression caused by interference and the use detection was reduced five thousand times (1.0 μg/mL), of chromatography separations to targeted applications showing an R2 > 0.999 that is another indicator of the re- (Figure 2(h)). .e interface has only a few centimeters in liability of this technique, and the possibility to be imple- length and operates in seconds; besides, it can be adapted to mented in routine analyses [45]. In the same manner, any MS system using atmospheric-pressure ionization- simultaneous analyses of methamphetamines, cocaine, targeted applications. In this analysis, a miniature differen- morphine, and Δ9-tetrahydrocannabinol were performed in tial ion mobility filter is used and placed in front of the en- a single blood spot by PS-MS in only 2 minutes, with trance of the mass spectrometer, and a solution of 10 ng/mL of minimal sample preparation through the extraction of the the sample was introduced using infusion introduction of ions compounds by solvents [46]. created by electrospray ionization source coupled with ion trap PS-MS has also been used in positive ionization mode to MS/MS. .is method allowed the characterization of samples obtain chemical profiles of illicit drugs such as blotter papers in 30 seconds, reducing case backlogs in the targeted analysis containing extracts and leaves of natural cannabinoids and of analytes of interest, showing the range of quantification of synthetic cannabinoids; here, 1 mg of blotter paper was used as around 0.01–10 ng⁄ uL of cocaine [52]. the PS ionization source. For this reason, the authors rec- Recently, a new method has been developed coupling ommend to be careful with the low sensitivity of this technique microfluidics with a miniature mass spectrometer in order to that was observed to possibly occur due to an ionic sup- quantify cocaine in urine samples. .is method is able to pression process, caused by the matrix effect (extracted im- deliver droplets of solvents to dried urine samples, sepa- purities from the surface of the blotter paper). .e results rating droplets of 80 µL of extracts, then performing splits provided a limit of detection of around 0.17 ppb [47]. from the hydrophilic dried urine zones and driving them to PS-MS/MS has been used in targeted drug screening the destination electrode for analysis. .e LOQ for cocaine using an Orbitrap QMS, one in positive mode and the other was 40 ng/mL [74]. in negative mode. In the positive ion mode, over 130 drugs Another recent method of direct analysis is touch spray and drug metabolites in postmortem samples were semi- (TS). In this technique, the sample is transferred to a sub- quantitatively determined, proving an adequate method in strate with subsequent ionization; in this manner, the postmortem analysis. In the analysis in negative mode, an substrate can serve both as the means for the sample col- ion-screening method was also developed for a small panel lection, ionization, and as straightforward handling analysis of barbiturates and structural analogs. .is method showed of either solid or liquid samples without pretreatment good qualitative agreement with LC–MS-MS; the true (Figure 2(i)). Using TS-MS coupled with MS/MS, drugs of positive rate of paper spray MS/MS was 92%, and the true abuse like Δ9-tetrahydrocannabinol and buprenorphine negative rate was over 98%. .is result shows that this were identified in spiked oral fluid using medical swabs technique possesses the necessary potential for acidic drug directly, providing limits of detection of around 50 ng/mL, detection and screening without sample preparation; how- which are sought by international forensic and toxicological ever, the authors did not present a list of possible in- societies. .is adaptation of medical swabs for TS-MS terferences during the analysis [48]. analysis allows noninvasive and direct sampling of neat Most recently, PS-MS has been used modifying the paper oral fluids; however, the authors affirm that the drying step through molecularly imprinted polymers (MIP) to create represents the most time-consuming part of the analytical a specific site for cocaine analysis in 1 mL of the oral fluid. In protocol, but the potential of the technique is high in terms this case, the PS was set by holding the membrane connected of specificity, selectivity, and sensitivity [9]. directly to the outlet probe of the ESI with a 0.5 mm wire More recently, laser diode thermal desorption (LDTD) using an alligator-type clip and applying a voltage of 4 V, and atmospheric solids analysis probe (ASAP) have been obtaining an LOQ of 100 ng/mL, and becoming a promising coupled with HRMS using APCI ionization in order to method to analyze cocaine [49]. generate high-quality data from multiple samples with none High-performance ion mobility spectrometry (HPIMS) or minimal sample preparation, with the purpose of identi- has been used along with electrospray ionization to detect fying synthetic cannabinoids/cathinones through full-MS and codeine and morphine in urine samples without extra MS/MS experiments. In ASAP, a melting-point capillary tube sample pretreatment (Figure 2(g)). However, issues of is used to introduce the sample into a stream of heated ni- charge suppression in the presence of drug mixtures trogen gas, which results in the sample being desorbed from 14 Journal of Analytical Methods in Chemistry the capillary [53], and the desorbed sample is then ionized by [3] H. H. Lee, J. F. Lee, S. Y. Lin, and B. H. Chen, “Simultaneous a corona discharge needle. During ASAP-MS analysis, it was identification of abused drugs, benzodiazepines, and new possible to examine solid and liquid samples transferred to the psychoactive substances in urine by liquid chromatography capillary surface (Figure 2(j)); whereas in the LDTD-MS tandem mass spectrometry,” Kaohsiung Journal of Medical analysis, the samples were extracted by a solvent. .is Sciences, vol. 32, no. 3, pp. 118–127, 2016. method uses a specially designed 96-well plate with stainless [4] S. Gwak and J. R. Almirall, “Rapid screening of 35 new psychoactive substances by ion mobility spectrometry (IMS) alloy steel inserts, where the sample is thermally desorbed and direct analysis in real time (DART) coupled to quad- from the stainless steel by an infrared laser which forms rupole time-of-flight mass spectrometry (QTOF-MS),” Drug neutral gas-phase molecules [54] (Figure 2(k)). .ese gas- Testing and Analysis, vol. 7, no. 10, pp. 884–893, 2015. phase molecules are carried into the mass spectrometer inlet [5] C. Poole, Gas Chromatography, Elsevier, New York, NY, USA, by compressed air. Before they enter the mass spectrometer 1st edition, 2012. inlet, a corona discharge needle ionizes the neutral molecules. [6] D. Cappelle, H. Neels, M. Yegles et al., “Gas chromatographic .is LDTD-APCI-MS method results in a completely determination of ethyl glucuronide in hair: Comparison automated analysis with low sampling times. .e authors between tandem mass spectrometry and single quadrupole recommended the use of both methods of ambient ioni- mass spectrometry,” Forensic Science International, vol. 249, zation, which allow rapid experiments from a single sample pp. 20–24, 2015. introduction. However, when performing the optimization, [7] M. Chèze, A. Lenoan, M. Deveaux, and G. Pépin, “De- termination of ibogaine and noribogaine in biological fluids they verified that the simplicity of ASAP design allows it to and hair by LC-MS/MS after Tabernanthe iboga abuse. Iboga be easily switched between API techniques and possible alkaloids distribution in a drowning death case,” Forensic positive/negative switching for a single sample introduction, Science International, vol. 176, no. 1, pp. 58–66, 2008. which provides many possibilities of optimization during [8] M. Domin and R. Cody, Ambient Ionization Mass Spec- the analysis. More studies in this field are required, especially trometry, Royal Society of Chemistry, Cambridge, UK, 2014. in possible interference of suppression of ionization [75]. [9] V. Pirro, A. K. Jarmusch, M. Vincenti, and R. G. Cooks, “Direct drug analysis from oral fluid using medical swab touch spray mass spectrometry,” Analytica Chimica Acta, 4. Conclusions vol. 861, pp. 47–54, 2015. [10] H. Wang, J. Liu, R. G. Cooks, and Z. Ouyang, “Paper spray for Mass spectrometry is the most important technique used in direct analysis of complex mixtures using mass spectrometry,” toxicological forensic analysis. MS coupled with chromatog- Angewandte Chemie, vol. 122, no. 5, pp. 889–892, 2010. raphy are the preferred techniques to identify new drugs or [11] L. Habala, J. Valentova,´ I. Pechova,´ M. Fuknova,´ and metabolites through screening analysis, providing excellent F. Dev´ınsky, “DART–LTQ ORBITRAP as an expedient tool results in limit of detection, precision, accuracy, and sensi- for the identification of synthetic cannabinoids,” Legal tivity, although it may be a time-consuming process. Direct Medicine, vol. 20, pp. 27–31, 2016. techniques with MS (with less sample preparation) are more [12] M. Carson and S. Kerrigan, “Quantification of suvorexant in likely to be used in target analysis or in routine qualitative urine using gas chromatography/mass spectrometry,” Journal analysis. However, sample complexity complicates the iden- of Chromatography B, vol. 1040, pp. 289–294, 2017. tification among compounds with similar fragmentation [13] J. Greaves and J. Roboz, Mass Spectrometry for the Novice, patterns, along with the problems caused by ionization CRC Press, Boca Raton, FL, USA, 2008. chemical suppression. As a result, recent developments in MS [14] Y.-H. Wu, K. Lin, S.-C. Chen, and Y.-Z. Chang, “Integration of GC/EI-MS and GC/NCI-MS for simultaneous quantitative are concerned with the necessity of creating new software in determination of opiates, amphetamines, MDMA, ketamine, order to help improve simplicity and robustness in the and metabolites in human hair,” Journal of Chromatography identification of drugs. .ere is a growing necessity to develop B, vol. 870, no. 2, pp. 192–202, 2008. more innovative methodologies to reduce time consumption [15] A. Amirav, A. Gordin, M. Poliak, and A. B. Fialkov, “Gas in the analyses, enhance sensitivity, and finally move forward chromatography-mass spectrometry with supersonic molec- towards greener chemistry. ular beams,” Journal of Mass Spectrometry, vol. 43, no. 2, pp. 141–163, 2008. [16] T. Alon and A. Amirav, “How enhanced molecular ions in Conflicts of Interest Cold EI improve compound identification by the NIST library,” Rapid Communications in Mass Spectrometry, vol. 29, .e authors declare that there are no conflicts of interest no. 23, pp. 2287–2292, 2015. regarding the publication of this paper. [17] M. Conti, V. Tazzari, M. Bertona, M. Brambilla, and P. Brambilla, “Surface-activated chemical ionization combined References with electrospray ionization and mass spectrometry for the analysis of cannabinoids in biological samples. Part I: analysis of [1] C. Moore, L. Marinetti, C. Coulter, and K. Crompton, 11-nor-9-carboxytetrahydro-cannabinol,” Rapid Communica- “Analysis of pain management drugs, specifically fentanyl, in tions in Mass Spectrometry, vol. 25, no. 11, pp. 1552–1558, 2011. hair: application to forensic specimens,” Forensic Science [18] J. Liang, W.-y. Wu, G.-x. Sun et al., “A dynamic multiple International, vol. 176, no. 1, pp. 47–50, 2008. reaction monitoring method for the multiple components [2] O. Beck, “Exhaled breath for drugs of abuse testing—evalu- quantification of complex traditional Chinese medicine ation in criminal justice settings,” Science and Justice, vol. 54, preparations: Niuhuang Shangqing pill as an example,” no. 1, pp. 57–60, 2014. Journal of Chromatography A, vol. 1294, pp. 58–69, 2013. Journal of Analytical Methods in Chemistry 15

[19] M. C. Sampedro, N. Unceta, A. Gomez-Caballero´ et al., [32] M. K. McGonigal, J. A. Wilhide, P. B. Smith, N. M. Elliott, and “Screening and quantification of antipsychotic drugs in hu- F. L. Dorman, “Analysis of synthetic phenethylamine street man brain tissue by liquid chromatography-tandem mass drugs using direct sample analysis coupled to accurate mass spectrometry: application to postmortem diagnostics of fo- time of flight mass spectrometry,” Forensic Science In- rensic interest,” Forensic Science International, vol. 219, ternational, vol. 275, pp. 83–89, 2017. no. 1–3, pp. 172–178, 2012. [33] J. L. Poklis, C. E. Wolf, O. I. ElJordi, K. Liu, S. Zhang, and [20] I. Shah, A. Petroczi, M. Uvacsek, M. Ranky,´ and A. Poklis, “Analysis of the first- and second-generation raving D. P. Naughton, “Hair-based rapid analyses for multiple drugs dragon novelty bath salts containing methylone and pente- in forensics and doping: application of dynamic multiple drone,” Journal of Forensic Sciences, vol. 60, pp. S234–S240, reaction monitoring with LC-MS/MS,” Chemistry Central 2015. Journal, vol. 8, no. 1, p. 73, 2014. [34] K. M. Evans-Nguyen, T. L. Hargraves, and A. N. Quinto, [21] N. M. Suni, P. Lindfors, O. Laine et al., “Matrix effect in the “Immunoaffinity nanogold coupled with direct analysis in real analysis of drugs of abuse from urine with desorption at- time (DART) mass spectrometry for analytical toxicology,” mospheric pressure photoionization-mass spectrometry Analytical Methods, vol. 9, no. 34, pp. 4954–4957, 2017. (DAPPI-MS) and desorption electrospray ionization-mass [35] W. F. Duvivier, T. A. van Beek, and M. W. F. Nielen, “Critical spectrometry (DESI-MS),” Analytica Chimica Acta, vol. 699, comparison of mass analyzers for forensic hair analysis by no. 1, pp. 73–80, 2011. ambient ionization mass spectrometry,” Rapid Communica- [22] J. .unig, L. Flø, S. Pedersen-Bjergaard, S. H. Hansen, and tions in Mass Spectrometry, vol. 30, no. 21, pp. 2331–2340, C. Janfelt, “Liquid-phase microextraction and desorption 2016. electrospray ionization mass spectrometry for identification [36] A. U. Jackson, J. F. Garcia-Reyes, J. D. Harper et al., “Analysis and quantification of basic drugs in human urine,” Rapid of drugs of abuse in biofluids by low temperature plasma Communications in Mass Spectrometry, vol. 26, no. 2, (LTP) ionization mass spectrometry,” Analyst, vol. 135, no. 5, pp. 133–140, 2012. p. 927, 2010. [23] Z. Lin, S. Zhang, M. Zhao, C. Yang, D. Chen, and X. Zhang, [37] T. Porta, C. Grivet, T. Kraemer, E. Varesio, and “Rapid screening of clenbuterol in urine samples by de- G. Hopfgartner, “Single hair cocaine consumption monitoring by mass spectrometric imaging,” Analytical Chemistry, sorption electrospray ionization tandem mass spectrometry,” vol. 83, no. 11, pp. 4266–4272, 2011. Rapid Communications in Mass Spectrometry, vol. 22, no. 12, [38] B. Flinders, E. Beasley, R. M. Verlaan et al., “Optimization of pp. 1882–1888, 2008. sample preparation and instrumental parameters for the rapid [24] T. J. Kauppila, A. Flink, M. Haapala et al., “Desorption at- analysis of drugs of abuse in hair samples by MALDI-MS/MS mospheric pressure photoionization–mass spectrometry in imaging,” Journal of :e American Society for Mass Spec- routine analysis of confiscated drugs,” Forensic Science In- trometry, vol. 28, no. 11, pp. 2462–2468, 2017. ternational, vol. 210, no. 1–3, pp. 206–212, 2011. [39] K. M. Ostermann, A. Luf, N. M. Lutsch et al., “MALDI [25] T. J. Kauppila, V. Arvola, M. Haapala et al., “Direct analysis of orbitrap mass spectrometry for fast and simplified analysis of illicit drugs by desorption atmospheric pressure photoioni- novel street and designer drugs,” Clinica Chimica Acta, zation,” Rapid Communications in Mass Spectrometry, vol. 22, vol. 433, pp. 254–258, 2014. no. 7, pp. 979–985, 2008. [40] H. Wang and Y. Wang, “Matrix-assisted laser desorption/ [26] R. B. Cody and J. A. Larame, “Versatile new ion source for the ionization mass spectrometric imaging for the rapid seg- analysis of materials in open air under ambient conditions,” mental analysis of methamphetamine in a single hair using Analytical Chemistry, vol. 77, no. 8, pp. 2297–2302, 2005. umbelliferone as a matrix,” Analytica Chimica Acta, vol. 975, [27] E. S. Chernetsova and G. E. Morlock, “Determination of drugs pp. 42–51, 2017. and drug-like compounds in different samples with direct [41] A. Kernalleguen,´ R. Steinhoff, S. Bachler et al., “High- analysis in real time mass spectrometry,” Mass Spectrometry throughput monitoring of cocaine and its metabolites in Reviews, vol. 35, no. 5, pp. 875–883, 2011. hair using microarrays for mass spectrometry and matrix- [28] R. A. Musah, M. A. Domin, M. A. Walling, and assisted laser desorption/ionization-tandem mass spectrom- J. R. E. Shepard, “Rapid identification of synthetic cannabi- etry,” Analytical Chemistry, vol. 90, no. 3, pp. 2302–2309, noids in herbal samples via direct analysis in real time mass 2018. spectrometry,” Rapid Communications in Mass Spectrometry, [42] E. Cuypers et al., “Article a closer look into the consequences vol. 26, no. 9, pp. 1109–1114, 2012. of decontamination procedures in forensic hair analysis using [29] R. A. Musah, M. A. Domin, R. B. Cody, A. D. Lesiak, A. John MetA-SIMS analysis,” Analytical Chemistry, 2016. Dane, and J. R. E. Shepard, “Direct analysis in real time mass [43] E. Cuypers, B. Flinders, C. M. Boone et al., “Consequences of spectrometry with collision-induced dissociation for struc- decontamination procedures in forensic hair analysis using tural analysis of synthetic cannabinoids,” Rapid Communi- metal-assisted secondary ion mass spectrometry analysis,” cations in Mass Spectrometry, vol. 26, no. 19, pp. 2335–2342, Analytical Chemistry, vol. 88, no. 6, pp. 3091–3097, 2016. 2012. [44] J. A. Michely, M. R. Meyer, and H. H. Maurer, “Paper spray [30] W. C. Samms, Y. J. Jiang, M. D. Dixon, S. S. Houck, and ionization coupled to high resolution tandem mass spec- A. Mozayani, “Analysis of alprazolam by DART-TOF mass trometry for comprehensive urine drug testing in comparison spectrometry in counterfeit and routine drug identification to liquid chromatography-coupled techniques after urine cases,” Journal of Forensic Sciences, vol. 56, no. 4, pp. 993–998, precipitation or dried urine spot workup,” Analytical 2011. Chemistry, vol. 89, no. 21, pp. 11779–11786, 2017. [31] A. H. Grange and G. W. Sovocool, “Detection of illicit drugs [45] T. C. De Carvalho, F. Tosato, L. M. Souza et al., “.in layer on surfaces using direct analysis in real time (DART) time-of- chromatography coupled to paper spray ionization mass flight mass spectrometry,” Rapid Communications in Mass spectrometry for cocaine and its adulterants analysis,” Fo- Spectrometry, vol. 25, no. 9, pp. 1271–1281, 2011. rensic Science International, vol. 262, pp. 56–65, 2016. 16 Journal of Analytical Methods in Chemistry

[46] R. D. Espy, S. F. Teunissen, N. E. Manicke et al., “Paper spray [60] R. Andrews and S. Paterson, “A validated method for the and extraction spray mass spectrometry for the direct and analysis of cannabinoids in post-mortem blood using liquid– simultaneous quantification of eight drugs of abuse in whole liquid extraction and two-dimensional gas chromatography– blood,” Analytical Chemistry, vol. 86, no. 15, pp. 7712–7718, mass spectrometry,” Forensic Science International, vol. 222, 2014. no. 1–3, pp. 111–117, 2012. [47] E. Domingos, T. C. de Carvalho, I. Pereira et al., “Paper spray [61] G. Milman, A. J. Barnes, R. H. Lowe, and M. A. Huestis, ionization mass spectrometry applied to forensic “Simultaneous quantification of cannabinoids and metabo- chemistry–drugs of abuse, inks and questioned documents,” lites in oral fluid by two-dimensional gas chromatography Analytical Methods, vol. 9, no. 30, pp. 4400–4409, 2017. mass spectrometry,” Journal of Chromatography A, vol. 1217, [48] J. McKenna, R. Jett, K. Shanks, and N. E. Manicke, “Toxi- no. 9, pp. 1513–1521, 2010. cological drug screening using paper spray high-resolution [62] B. Guthery, T. Bassindale, A. Bassindale, C. T. Pillinger, and tandem mass spectrometry (HR-MS/MS),” Journal of Ana- G. H. Morgan, “Qualitative drug analysis of hair extracts by lytical Toxicology, vol. 42, no. 5, pp. 300–310, 2018. comprehensive two-dimensional gas chromatography/time- [49] L. S. Tavares, T. C. Carvalho, W. Romão, B. G. Vaz, and of-flight mass spectrometry,” Journal of Chromatography A, A. R. Chaves, “Paper spray tandem mass spectrometry based vol. 1217, no. 26, pp. 4402–4410, 2010. on molecularly imprinted polymer substrate for cocaine [63] A. Salomone, E. Gerace, F. D’Urso, D. Di Corcia, and analysis in oral fluid,” Journal of :e American Society for M. Vincenti, “Simultaneous analysis of several synthetic Mass Spectrometry, vol. 29, no. 3, pp. 566–572, 2017. cannabinoids, THC, CBD and CBN, in hair by ultra-high [50] A. J. Midey, A. Patel, C. Moraff, C. A. Krueger, and C. Wu, performance liquid chromatography tandem mass spec- “Improved detection of drugs of abuse using high- trometry. Method validation and application to real samples,” performance ion mobility spectrometry with electrospray Journal of Mass Spectrometry, vol. 47, no. 5, pp. 604–610, 2012. ionization (ESI-HPIMS) for urine matrices,” Talanta, vol. 116, [64] M. Sundström,A. Pelander, V. Angerer, M. Hutter, S. Kneisel, pp. 77–83, 2013. and I. Ojanpera,¨ “A high-sensitivity ultra-high performance [51] T. Gabowitcz, D. Ridjosic, and S. Nacson, Ion Mobility liquid chromatography/high- resolution time-of-flight mass Spectrometer Having Improved Sample Receiving Device, US spectrometry (UHPLC-HR-TOFMS) method for screening 2008/0101995 A1, 2008. synthetic cannabinoids and other drugs of abuse in urine,” [52] A. B. Hall, S. L. Coy, E. G. Nazarov, and P. Vouros, “Rapid Analytical and Bioanalytical Chemistry, vol. 405, no. 26, separation and characterization of cocaine and cocaine cut- pp. 8463–8474, 2013. [65] E. Partridge, S. Trobbiani, P. Stockham, T. Scott, and ting agents by differential mobility spectrometry-mass spec- C. Kostakis, “A validated method for the screening of 320 trometry,” Journal of Forensic Sciences, vol. 57, no. 3, forensically significant compounds in blood by LC/QTOF, pp. 750–756, 2012. with simultaneous quantification of selected compounds,” [53] E. Jagerdeo, J. A. Clark, J. N. Leibowitz, and L. J. Reda, “Rapid Journal of Analytical Toxicology, vol. 42, no. 4, pp. 220–231, analysis of forensic samples using an atmospheric solid 2018. analysis probe interfaced to a linear ion trap mass spec- [66] I.-T. Wang, Y.-T. Feng, and C.-Y. Chen, “Determination of 17 trometer,” Rapid Communications in Mass Spectrometry, illicit drugs in oral fluid using isotope dilution ultra-high vol. 29, no. 2, pp. 205–212, 2015. performance liquid chromatography/tandem mass spec- [54] J. Wu, C. S. Hughes, P. Picard et al., “High-throughput cyto- trometry with three atmospheric pressure ionizations,” chrome P450 inhibition assays using laser diode thermal Journal of Chromatography B, vol. 878, no. 30, pp. 3095–3105, desorption-atmospheric pressure chemical ionization-tandem 2010. mass spectrometry,” Analytical Chemistry, vol. 79, no. 12, [67] A. G. Helfer, J. A. Michely, A. A. Weber, M. R. Meyer, and pp. 4657–4665, 2007. H. H. Maurer, “Orbitrap technology for comprehensive [55] A. Amirav, “Fast heroin and cocaine analysis by GC–MS with metabolite-based liquid chromatographic–high resolution- cold EI: the important role of flow programming,” Chro- tandem mass spectrometric urine drug screening–exemplified matographia, vol. 80, no. 2, pp. 295–300, 2017. for cardiovascular drugs,” Analytica Chimica Acta, vol. 891, [56] M. K. Wozniak,´ M. Wiergowski, J. Aszyk, P. Kubica, pp. 221–233, 2015. J. Namie´snik, and M. Biziuk, “Application of gas [68] T. R. Fiorentin, F. B. D’Avila, E. Comiran et al., “Simultaneous chromatography–tandem mass spectrometry for the de- determination of cocaine/crack and its metabolites in oral termination of amphetamine-type stimulants in blood and fluid, urine and plasma by liquid chromatography-mass urine,” Journal of Pharmaceutical and Biomedical Analysis, spectrometry and its application in drug users,” Journal of vol. 148, pp. 58–64, 2018. Pharmacological and Toxicological Methods, vol. 86, pp. 60– [57] F. Versace, F. Sporkert, P. Mangin, and C. Staub, “Rapid 66, 2017. sample pre-treatment prior to GC–MS and GC–MS/MS [69] A. E. Steuer, M. Poetzsch, L. Stock et al., “Development and urinary toxicological screening,” Talanta, vol. 101, pp. 299– validation of an ultra-fast and sensitive microflow liquid 306, 2012. chromatography-tandem mass spectrometry (MFLC-MS/ [58] E. S. Em´ıdio, V. de Menezes Prata, and H. S. Dorea,´ “Vali- MS) method for quantification of LSD and its metabolites dation of an analytical method for analysis of cannabinoids in in plasma and application to a controlled LSD administration hair by headspace solid-phase microextraction and gas study in huma,” Drug Testing and Analysis, vol. 9, no. 5, chromatography–ion trap tandem mass spectrometry,” pp. 788–797, 2017. Analytica Chimica Acta, vol. 670, no. 1-2, pp. 63–71, 2010. [70] F. Xian, C. L. Hendrickson, and A. G. Marshall, “High res- [59] K. Zaitsu, H. Miyagawa, Y. Sakamoto et al., “Mass spectrometric olution mass spectrometry,” Analytical Chemistry, vol. 84, differentiation of the isomers of mono-methoxyethylamphetamines no. 2, pp. 708–719, 2012. and mono-methoxydimethylamphetamines by GC–EI–MS–MS,” [71] D. Pasin, A. Cawley, S. Bidny, and S. Fu, “Current applications Forensic Toxicology, vol. 31, no. 2, pp. 292–300, 2013. of high-resolution mass spectrometry for the analysis of new Journal of Analytical Methods in Chemistry 17

psychoactive substances: a critical review,” Analytical and Bioanalytical Chemistry, vol. 409, no. 25, pp. 5821–5836, 2017. [72] J. D. Harper, N. A. Charipar, C. C. Mulligan, X. Zhang, R. G. Cooks, and Z. Ouyang, “Low-temperature plasma probe for ambient desorption ionization,” Analytical Chemistry, vol. 80, no. 23, pp. 9097–9104, 2008. [73] J. M. Wiseman, B. Gologan, and R. G. Cooks, “Mass spectrometry sampling under ambient conditions with desorption electrospray ionization,” Science, vol. 306, no. 5695, pp. 471– 474, 2004. [74] A. E. Kirby, N. M. Lafreniere,` B. Seale, P. I. Hendricks, R. G. Cooks, and A. R. Wheeler, “Analysis on the go: quantitation of drugs of abuse in dried urine with digital microﬂuidics and miniature mass spectrometry,” Analytical Chemistry, vol. 86, no. 12, pp. 6121–6129, 2014. [75] E. Jagerdeo and A. Wriston, “Rapid analysis of forensic- related samples using two ambient ionization techniques coupled to high-resolution mass spectrometers,” Rapid Communications in Mass Spectrometry, vol. 31, no. 9, pp. 782–790, 2017. Hindawi Journal of Analytical Methods in Chemistry Volume 2018, Article ID 7019204, 9 pages https://doi.org/10.1155/2018/7019204

Research Article Establishing Analytical Performance Criteria for the Global Reconnaissance of Antibiotics and Other Pharmaceutical Residues in the Aquatic Environment Using Liquid Chromatography-Tandem Mass Spectrometry

Luisa F. Angeles and Diana S. Aga

Department of Chemistry, e State University of New York, Buﬀalo, NY 14260, USA

Correspondence should be addressed to Diana S. Aga; dianaaga@buﬀalo.edu

Received 10 March 2018; Accepted 26 April 2018; Published 4 June 2018

Academic Editor: Veronica Termopoli

Copyright © 2018 Luisa F. Angeles and Diana S. Aga. *is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

*e occurrence of antibiotics in the environment from discharges of wastewater treatment plants (WWTPs) and from the land application of antibiotic-laden manure from animal agriculture is a critical global issue because these residues have been associated with the increased emergence of antibiotic resistance in the environment. In addition, other classes of pharmaceuticals and personal care products (PPCPs) have been found in effluents of municipal WWTPs, many of which persist in the receiving environments. Analysis of antibiotics by liquid chromatography-tandem mass spectrometry (LC-MS/MS) in samples from different countries presents unique challenges that should be considered, from ion suppression due to matrix effects, to lack of available stable isotopically labeled standards for accurate quantification. Understanding the caveats of LC-MS/MS is important for assessing samples with varying matrix complexity. Ion ratios between quantifying and qualifying ions have been used for quality assurance purposes; however, there is limited information regarding the significance of setting criteria for acceptable variabilities in their values in the literature. Upon investigation of 30 pharmaceuticals in WWTP influent and effluent samples, and in receiving surface water samples downstream and upstream of the WWTP, it was found that ion ratios have higher variabilities at lower concentrations in highly complex matrices, and the extent of variability may be exacerbated by the physicochemical properties of the analytes. In setting the acceptable ion ratio criterion, the overall mean, which was obtained by taking the average of the ion ratios at all concentrations (1.56 to 100 ppb), was used. *en, for many of the target analytes included in this study, the tolerance range was set at 40% for WWTP influent samples and 30% for WWTP effluent, upstream, and downstream samples. A separate tolerance range of 80% was set for tetracyclines and quinolones, which showed higher variations in the ion ratios compared to the other analytes.

1. Introduction WWTPs in different geographical regions has been docu- mented, with concentrations reported as high as about In recent years, studies have reported the occurrence of 125 µg/L [10]. In Germany, the environmental concentration pharmaceuticals and personal care products (PPCPs), in- in municipal sewage that comes from the discharge of an- cluding antibiotics and selective serotonin reuptake in- tibiotics from hospitals and households is predicted to be hibitors (SSRIs), in the environment [1–6]. *ese drugs are about 71 mg/L annually [9]. *e presence of high levels of being released through different routes, such as discharges pharmaceuticals in the environment has a wide range of from wastewater treatment plant (WWTP) effluents to ecological effects; for instance, antibiotics may contribute to surface water, where hospitals and private households the development of antibiotic resistance in bacteria due to contribute a large volume of antibiotics and other phar- selective pressure, which is a threat to global health maceuticals [2, 7–9]. *e presence of PPCPs in effluents of [9, 11, 12]. 2 Journal of Analytical Methods in Chemistry

Analysis of PPCPs in environmental samples is typically of the sample matrix. *is variability is due to differences in performed using liquid chromatography-tandem mass the ionization behavior of analytes and the extent of matrix spectrometry (LC-MS/MS) to quantify pharmaceutical effects. It is not unexpected to observe different effects on the concentrations based on triple quadrupole MS [3, 6, 13]. *e ion ratios of the analytes in wastewater and surface water high selectivity and sensitivity obtained using triple quad- matrices because the composition of the interferences in rupole MS is achieved when performing selected reaction environmental samples is different relative to biological monitoring (SRM), where a precursor ion is isolated from samples. the first quadrupole and fragmented in the collision cell, Establishing performance criteria is important because it followed by isolating selected product ions in the third minimizes the occurrence of false-positive and false-negative quadrupole. However, despite this high selectivity, there is detections. In fact, a doubling of false-positive detections was still a possibility that a compound other than the target reported without the application of the ion ratio criterion in analyte will produce a signal that has a similar m/z value to the analysis of veterinary drugs in the muscle, urine, milk, either the qualifying ion or the quantifying ion at the same and liver [14]. Most published and existing methods do not retention time [14], resulting in a significant deviation in the mention the use of any ion ratio criteria [3, 6, 13]. In the US expected ion ratio for the selected fragment ions being Environmental Protection Agency (EPA) Method 1694 for monitored by the two SRM transitions. the determination of PPCPs in environmental samples by In order to confirm the presence of a compound, the LC-MS/MS, the presence of a compound in a sample extract chromatographic peak must have both the quantitative and is confirmed when the signal-to-noise ratio (S/N) of the qualitative ion transitions with retention times matching those fragment ion of the compound is greater than or equal to 2.5 of the standard analyte. In addition, the ion ratio of the two and its retention time is within ±15 seconds of the cali- SRM transitions has been used as an additional confirmation bration verification standard. If these criteria are not met, criterion, as stated in some legal documents from different then an experienced analyst must confirm the presence or organizations such as the European Union (EU) and the US absence of a compound [22]. Additionally, in the EPA Food and Drug Administration (US FDA), which provide Method 542, which is for the analysis of PPCPs in drinking guidelines for the analysis of official samples [15–19]. Having water, the acceptable retention time window for the com- this additional criterion is important since LC-MS/MS has now pounds in a sample is within 3 standard deviations for become the mandatory technique for the analysis of official a series of injections. Quality control for this method in- samples that are used for establishing legal policies [15–19]. volves the confirmation of the presence of the quantifying Monitoring the ion ratios will provide improved confidence in ion of the internal standard and requires that it must be reporting analyte concentrations, avoiding false positives and within ±50% of the average area measured in the initial false negatives, which have been reported in the literature [14]. calibration [23]. No criteria regarding the ion ratios have Different legal guidelines are currently available from been mentioned in both EPA methods. *e absence of the United Nations (UN), the EU, and the United States of quality control measures in published methods may be due America (USA). *e UN set the ion ratio tolerance to be to the lack of suitable guidelines in the literature. In order to ±20% [16] for the testing of illicit drugs in seized materials determine an appropriate tolerance value for the ion ratios, and biological specimens. *e European Commission variabilities resulting from the physicochemical nature of the Decision (2002/657/EC) requires a tolerance of ±20% to analytes should be investigated at high and low concen- ±50% for the ion ratio, depending on the ion intensities trations. *e variability in the signal intensities of the [17], for analytical methods that are used for the testing of qualifier ions is expected to be more significant than that of official samples in control laboratories. *e European the quantifier ions because of the relatively lower signals for Workplace Drug Testing Society sets it at ±20% [19]. *e the qualifier ions. US Department of Agriculture requires a ±20% tolerance *e aim of this study is to validate and provide guidelines in the ratio of the ion transitions [18], while the US FDA on the use of ion ratios as a criterion for quality control in sets an ion ratio tolerance of ±20% and ±30% if 2 and 3 reporting concentrations of PPCPs in wastewater and sur- diagnostic ions are being monitored, respectively [15]. *e face water samples with varying complexity. To achieve this weakness of these guidelines, however, is that they are not goal, the ion ratios of 30 PPCPs in different matrices were based on experimental data and are arbitrarily assigned. determined at different concentrations in order to determine Recent studies [20, 21] have been published on per- a tolerance value that is sufficient to eliminate false positives formance criteria for the analyses of pesticides in fruits and and false negatives. *e matrices studied were WWTP in- vegetables and veterinary drugs in biological matrices. For fluent and effluent samples and surface water samples from pesticides, a tolerance range of ±20% was established for all upstream and downstream of the WWTP discharge point compounds at all concentrations, except when one or both collected from the US, Sweden, Switzerland, Hong Kong, product ions have an S/N of 3–15, in which case, a range of and the Philippines, allowing the set tolerance levels to be ±45% was set. For veterinary drugs, a fixed tolerance range of robust, given that the composition of water samples varies ±50% for all the compounds at all concentrations was set significantly in different parts of the world. *e data ob- after evaluation of the ion ratios in different matrices such as tained from these analyses were the basis for the con- muscle, urine, milk, and liver [20, 21]. However, these struction of a more accurate and reliable ion ratio criterion tolerance values cannot be used for PPCPs because the which takes into account the differences in the properties of variability of ion ratios differs per compound and the nature compounds at different concentrations. Journal of Analytical Methods in Chemistry 3

2. Materials and Methods Table 1: Target pharmaceuticals used for establishing the ion ratio criterion. Acetaminophen (ACT), acetylsulfamethoxazole (ASMX), azi- Class Compound thromycin (AZI), caffeine (CAF), carbamazepine (CBZ), clarithromycin (CLA), enrofloxacin (ENRO), erythromycin Antibiotics Anhydroerythromycin (ERY), iopamidol (IOPA), norfloxacin (NOR), oxytetracycline Macrolides Azithromycin (OTC), sarafloxacin (SARA), sulfachloropyridazine (SCP), Clarithromycin sulfadiazine (SPD), sulfadimethoxine (SDM), sulfamerazine Ciprofloxacin (SMR), sulfameter (SMT), sulfamethazine (SMZ), sulfame- Enrofloxacin Quinolones thizole (SMI), sulfamethoxazole (SMX), tetracycline (TC), and Norfloxacin trimethoprim (TMP) were purchased from Sigma-Aldrich. Sarafloxacin Ciprofloxacin (CIP) and diclofenac (DIC) were obtained Acetylsulfamethoxazole from Cambridge Isotope Laboratories, Inc. (Tewksbury, MA). Sulfachloropyridazine Sulfathiazole was purchased from ICN Biomedicals, Inc. Sulfadiazine (Irvine, CA). Carbamazepine-d10 (d10-CBZ) was purchased Sulfadimethoxine Sulfamerazine from CDN Isotopes (Quebec, Canada). Chlortetracycline Sulfonamides (CTC) was obtained from Acros Organics (VWR International, Sulfamethazine Westchester, PA). Paroxetine maleate (PRX) and venlafaxine Sulfamethizole (VEN) were obtained from Cerilliant (Sigma-Aldrich, St Louis, Sulfamethoxazole MO). *e Barnstead NANOpure DIamond (Waltham, MA) Sulfamethoxydiazine ™ Sulfathiazole purification system was used to obtain 18.2 MΩ water. LC-MS grade methanol and acetonitrile were obtained from EMD Chlortetracycline Tetracyclines Oxytetracycline Millipore Corporation (Billerica, MA), and formic acid (88%) Tetracycline was purchased from Fisher Chemical (Pittsburgh, PA). Oasis ™ Other PPCPs HLB solid-phase extraction (SPE) cartridges were purchased Acetaminophen from Waters (Milford, MA). Caffeine Carbamazepine 2.1. Sample Preparation. Wastewater and surface water Diclofenac Iopamidol samples (0.5 L) were collected in amber glass bottles which Trimethoprim were pre-rinsed with 10% nitric acid. *e samples were Bupropion acidified to about pH 2.5 using 40% phosphoric acid and Paroxetine then passed through 0.45 µm glass microfiber filters to Sertraline remove microorganisms and particulate matter. *en, 2 mL Venlafaxine of Na2EDTA (5% w/v in water) was added to each sample. *e samples were then spiked with surrogate standards (50 µL of 1000 µg/L surrogate mix solution). a heated electrospray ionization (HESI) probe, operated *e samples were passed through Oasis HLB SPE car- under positive ionization mode. Timed-SRM mode transi- tridges (500 mg, 6 cc) for cleanup and concentration. *e SPE tion was performed, and the SRM transitions used for the cartridges were first conditioned with 6 mL acetonitrile, fol- compounds are shown in Table S1. lowed by 6 mL NANOpure water, before the water samples *e mobile phase used for the separation consisted of were loaded at a rate of approximately 3–5 mL/min. After aqueous 0.3% formic acid (A), and 75% methanol and 25% loading, the cartridges were dried by keeping them on the SPE acetonitrile (B). *e gradient began with 90% A and 10% B manifold with the vacuum on. *en, the SPE cartridges were for three minutes and is ramped up linearly to 100% B for wrapped in aluminum foil, stored in Ziploc® bags, and shipped 22 min; this condition was kept for 5 min before it was with ice to the University at Buffalo for elution and LC-MS/MS switched back to 90% A, where it was maintained for 14 min analysis. Once received, the samples were eluted using 8 mL of to allow for column equilibration. *e flow rate was set at ° acetonitrile and then dried under N2 gas at 35 C. *e samples 0.2 mL/min, and the total run time was 45 min. were then spiked with 100 ppb of the internal standard, *e spray setting used for the MS was as follows: spray carbamazepine-d10, in order to account for possible differ- voltage 3000 V, ion sweep gas pressure 0 arbitrary units, va- ences in measurements in-between injections due to variations porizer temperature 350°C, sheath gas pressure 40 arbitrary caused by the instrument. units (N2), auxiliary gas pressure 35 arbitrary units (N2), capillary temperature 325°C, collision gas pressure 1.5 mTorr (Ar), cycle time 0.300 s, and Q1 peak width 0.70 FWHM. 2.2. LC-MS/MS Analysis. A Waters Cortecs™ C18+ column (Milford, MA) with dimensions 2.1 × 150 mm and 2.7 µm particle size was used for the separation of the 30 PPCPs. 2.3. Design of the Study Analysis was performed using an Agilent 1200 LC system (Palo Alto, CA) and a *ermo Scientific TSQ Quantum 2.3.1. Assessment of Ion Ratio Behavior of PPCPs across Ultra triple quadrupole MS (Waltham, MA) equipped with Varying Concentrations. A total of 30 PPCPs were studied 4 Journal of Analytical Methods in Chemistry for the development of an ion ratio criterion (Table 1). A 100 mixture of all the native PPCP standards was prepared using 80 the starting mobile phase of the LC-MS/MS method as the 60 solvent. An initial solution of 100 ppb (µg/L) was made, and 40 then it was serially diluted to obtain mixtures with con- 20 centrations of 50, 25, 12.5, 6.25, 3.13, and 1.56 ppb. *ese standards were analyzed by LC-MS/MS, with nine replicates 0 for each concentration, to obtain the areas of both the –20 quantifying and qualifying ions. *e ion ratios were cal- –40 culated by dividing the area of the quantifying ion by the Deviation from mean (%) mean from Deviation –60 area of the qualifying ion for each analyte. *e average ion –80 ratio and the deviations from the average value for the ion –100 ratios were calculated at all concentrations for each 1.56 6.253.13 12.5 25.0 50.0 100 compound. Concentration (ppb)

Figure 1: Deviation of ion ratios from the overall mean across 2.3.2. Assessment of Ion Ratio Behavior of Pharmaceuticals in different concentrations of 23 PPCPs without tetracyclines and the Matrix. Samples from WWTP influents and effluents quinolones. and from receiving surface waters upstream and downstream of the WWTPs were collected from selected sites in five countries: Central, Hong Kong; Manila, Philippines; 140 Vastergotland, Sweden; Zurich, Switzerland; and Virginia, 120 US. *e samples from the Philippines were collected in 100 December 2016, while the others were collected in June or 80 July 2016. *e exact names and locations of the WWTPs 60 cannot be disclosed as part of the agreement with the 40 WWTP operations. A total of 19 samples were each spiked 20 with 1.56 ppb, 12.5 ppb, 25 ppb, and 100 ppb of the native 0 standard mix and were analyzed by LC-MS/MS to determine –20 the mean ion ratios and standard deviations from the mean Deviation from mean (%) mean from Deviation –40 for each compound. *e variabilities of the ion ratios in the –60 different sample matrices were then evaluated and compared –80 with the values observed in the standards. 1.56 3.13 6.25 12.5 25.0 50.0 100 Concentration (ppb)

2.3.3. Optimization of the Ion Ratio Criterion. *e proposed Figure 2: Deviation of ion ratios of tetracyclines and quinolones formula to be used in order to optimize the appropriate ion from the overall mean across different concentrations. ratio tolerance that will give the least false negative is a mean ion ratio and a tolerance range that will account for variations in the sample matrix. *is tolerance range should not 50.0, and 100 ppb in the LC-MS/MS. *e overall mean, be too wide so as to avoid having false positives. To de- which is the average ion ratio of all nine replicates at all termine the optimum tolerance level, different values were concentrations, was obtained for each compound (Table S2). tested for all compounds; the same test was also used to *e relative percent deviation was then calculated by sub- determine whether a single tolerance range would be used tracting the overall mean from each of the data points and for all the matrices or if different ones should be used for then dividing by the overall mean. *ese values were then wastewater and for surface water. To check the appropri- plotted against the seven concentrations to see how the ion ateness of the selected tolerance values, the number of false ratios at each concentration vary from the overall mean of negatives will be determined using the water extracts spiked each compound, as shown in Figure 1. *e trend for all the with known amounts of standards. compounds is that the variation is highest at the lowest concentration. *e average relative standard deviation for all 3. Results and Discussion the compounds at 1.56 ppb was 18%, while that for compounds at 100 ppb was only 4%. *ese results indicate that 3.1. Ion Ratio Variability at Different Concentrations. A total the differences in the ion ratios at different concentrations of 30 PPCPs, which include 23 antibiotics, were studied for should be taken into account because if only one tolerance the development of an ion ratio criterion (Table 1). *e limit is applied across all concentrations, it is likely that false- classes of antibiotics that were included in this study were negative results will occur at low concentrations, especially sulfonamides, macrolides, quinolones, and tetracyclines. at concentrations between 1.56 to 12.5 ppb. First, the ion ratio behavior of the compounds was *e general trend in the ion ratios for 23 PPCPs is shown studied at different concentrations by analyzing nine rep- in Figure 1; a separate plot for tetracyclines and quinolones licates of the standard solutions of 1.56, 3.13, 6.25, 12.5, 25.0, was prepared (Figure 2) because the variabilities in the ion Journal of Analytical Methods in Chemistry 5

50 variations are higher in the wastewater as compared to those 45 of the surface water samples, with relative standard deviation 40 values of 13%, 11%, 10%, and 9%, for the influent, effluent, 35 30 downstream, and upstream samples, respectively. *is trend 25 was expected since the upstream and downstream samples 20 are less-complex matrices (lower organic matter content 15 than wastewater). It can be seen in the lowest concentration 10 studied (1.56 ppb) that the standards in the clean matrix Relative frequency (%) frequency Relative 5 varied more than the ones spiked in the samples, with 0 –90 –70 –50 –30 –10 10 30 50 70 90 relative standard deviations of 18% for the standards and Deviation from mean (%) 20%, 14%, 15%, and 12% for the influent, effluent, downstream, and upstream samples, respectively. *is is due to Quantitative ion the removal of 22 data points at 1.56 ppb because the Qualitative ion qualitative ions were no longer detected in the samples. Figure 3: Distribution of the deviation of the quantitative and qualitative ion areas from the mean. 3.3. Optimization of a Tolerance Range for the Ion Ratio ratios were notably higher in these classes of antibiotics than Criterion. *e formula for the ion ratio criterion that was the rest of the PPCPs. Data points at lower concentrations of used is a tolerance range from the mean of each standard some compounds were removed in cases where the quali- compound in the clean matrix. *is tolerance range should tative ion was not detectable. For example, oxytetracycline account for the deviations because of differences in con- was not detected below 25 ppb, chlortetracycline and tet- centrations and the matrix being analyzed. *e goal in racycline were not detected below 12.5 ppb, sarafloxacin and setting this range is to have the least number of false positives norfloxacin were not detected below 6.25 ppb, and enro- and false negatives. False negatives will occur when the ion floxacin was not detected at 1.56 ppb. *erefore, a separate ratio of analytes in spiked environmental samples does not chart (Figure 2) was created for tetracyclines and quinolones meet the tolerance criteria such that the analyte in question since they do not follow the same behavior as the other will be considered “nondetect.” On the contrary, one cannot pharmaceuticals. Based on these data, a separate ion ratio set a tolerance range too wide that will likely result in tolerance range is needed for tetracyclines and quinolones in a significant number of false positives. *erefore, a range order to capture the wide variations, without affecting the that will still capture all the variations at the 95% confidence other compounds. It can be observed that the variations in level in both matrices and at different concentrations is Figure 2 are lower at 1.56 ppb compared to the higher needed. concentrations, but this is because most of these compounds In order to provide an appropriate criterion, the overall were no longer detected at 1.56 ppb, and these data points mean, which is obtained by taking the average of the ion were removed in the chart. *e deviation from the mean ratios at all concentrations (1.56 to 100 ppb), will be used. reaches up to 120% at 12.5 ppb for tetracyclines and up to *is way, the variations of the ion ratios from low to high 50% for quinolones at 6.25 ppb. At the highest concentration concentrations will be taken into account. *e tolerance of 100 ppb, the deviations from the mean in both tetracy- range must then be optimized for the spiked matrices. clines and quinolones are at 40%, while those for the other Tolerance ranges from 10% to 50% were tested to see which PPCPs are only 20%. one will give the least number of false negatives for each of *e areas of both the quantitative and qualitative ions the water matrices (Table 2). *e overall false-negative rate is were investigated separately in order to identify which of the the weighted average of the percent false negatives for each two ions causes high variations. *e deviations in the areas matrix type. of these ions from the mean were calculated and compared Since the tetracyclines and quinolones were found to with each other. Since the distribution of variation of both have greater variations than the rest of the PPCPs as seen in the qualitative and quantitative areas is similar, as seen in Figure 2, a test was performed to check if tetracyclines and Figure 3, this means that both of them contribute equally to quinolones should use a different tolerance level than what is the variations, and the ion ratio deviations cannot be at- used for the other classes of PPCPs. Table 3 shows the tributed to just the quantitative or qualitative ion alone. percentage of false negatives in the samples when the tetracyclines and quinolones were removed. If tetracyclines and quinolones were included, a tolerance range of 50% would 3.2. Ion Ratio Variability in Wastewater and Surface Water give a false-negative result of ≤5%. If removed, a tolerance Matrices. A total of 19 different samples were spiked with range of 30% would be enough to give the same value of ≤5% the pharmaceutical standards at 4 concentrations: 1.56, 12.5, for false negatives. 25.0, and 100 ppb, in order to determine how the differences It is important to have a separate tolerance range for in the nature of the matrices influence the ion ratios. tetracyclines and quinolones because as seen in Table 2, Figure 4 shows how the ion ratios change in the influent, a tolerance range of 50% is needed in order to capture them effluent, upstream, and downstream water samples in at the 95% confidence level. *is value, however, would be comparison with the clean standard. It can be seen that the too high for the other PPCPs, where only 30% is required to 6 Journal of Analytical Methods in Chemistry

200 200

150 150

100 100

50 50

0 0

–50 –50 Deviation from mean (%) mean from Deviation Deviation from mean (%) mean from Deviation –100 –100 1.56 12.5 25.0 100 1.56 12.5 25.0 100 Concentration (ppb) Concentration (ppb) Standards Standards Spiked matrix (infuent) Spiked matrix (efuent)

(a) (b) 200 200

150 150

100 100

50 50

0 0

Figure 4: Comparison of ion ratios in spiked matrices and in clean standards. (a) WWTP inﬂuent samples; (b) WWTP eﬄuent samples; (c) upstream surface water samples; (d) downstream surface water samples.

Table 2: Percent false negatives in spiked environmental matrices at different tolerance ranges for all 30 PPCPs. Matrix Tolerance range Overall false negatives Influent Effluent Downstream Upstream ±10% 54% 49% 47% 42% 38% ±20% 26% 20% 16% 14% 16% ±30% 13% 9% 7% 6% 8% ±40% 8% 6% 4% 3% 6% ±50% 5% 4% 2% 2% 4%

Table 3: Percent false negatives in spiked environmental matrices at different tolerance ranges for 23 compounds without tetracyclines and quinolones. Matrix Tolerance range Overall false negatives Influent Effluent Downstream Upstream ±10% 48% 42% 41% 35% 31% ±20% 20% 13% 10% 8% 10% ±30% 8% 4% 2% 2% 4% ±40% 4% 2% 0% 0% 2% ±50% 2% 1% 0% 0% 1% have the same confidence level (Table 3). *erefore, a fixed below 5% for many PPCPs, as seen in Table 2, it was observed tolerance range of 50% for all compounds could potentially that the tetracyclines and quinolones still had very high values, result in high false negatives for tetracyclines and quinolones with ciprofloxacin having 96% false negatives (Table 4). and high false positives for the other PPCPs. Also, at the 50% For chlortetracycline, ciprofloxacin, enrofloxacin, nor- tolerance range, even if the overall false negatives were already floxacin, and tetracycline, the range needed to be from 70% Journal of Analytical Methods in Chemistry 7

Table 4: Percent false negatives for tetracyclines and quinolones at the 50% tolerance range in spiked environmental matrices. Matrix Compounds Overall false negatives Influent Effluent Downstream Upstream Chlortetracycline 10% 0% 0% 0% 2% Ciprofloxacin 96% 71% 33% 13% 61% Enrofloxacin 5% 0% 0% 0% 1% Norfloxacin 0% 4% 0% 0% 1% Oxytetracycline 0% 30% 29% 22% 21% Sarafloxacin 12% 16% 0% 20% 13% Tetracycline 11% 0% 0% 0% 3%

Table 5: Percent false negatives for tetracyclines and quinolones in spiked environmental matrices at the 80% tolerance range. Matrix Compounds Overall false negatives Influent Effluent Downstream Upstream Chlortetracycline 10% 0% 0% 0% 2% Ciprofloxacin 0% 0% 0% 0% 0% Enrofloxacin 0% 0% 0% 0% 0% Norfloxacin 0% 4% 0% 0% 1% Oxytetracycline 0% 0% 29% 0% 6% Sarafloxacin 6% 11% 0% 20% 10% Tetracycline 0% 0% 0% 0% 0%

Table 6: Results of the application of the ion ratio criterion in real wastewater and surface water samples. Matrix Tolerance range Total no. of detections No. of detections outside the range Influent ±40% 102 0 Effluent ±30% 90 2 Downstream ±30% 37 2 Upstream ±30% 39 1 A number of detections outside the range are data points that were considered positive detections but had ion ratios outside the set tolerance range. to 85% in order to have a false-negative rate of ≤5%. higher, 26%, 20%, 16%, and 14%, for the influent, effluent, Oxytetracycline and sarafloxacin, on the contrary, still have downstream, and upstream samples, respectively. *erefore, false negatives of up to 29% (downstream) and 20% (up- it is important to have a separate tolerance range for certain stream), respectively, at a tolerance range of 80% (Table 5). compounds in different environmental matrices. However, setting a wider range may result in greater prob- ability of false positives. *erefore, a tolerance range of 80% was set for the tetracyclines and quinolones, but it is rec- 3.4. Applying the Optimized Ion Ratio Criterion in Real Water ommended that other criteria such as retention time, peak Samples from around the World. *e optimized ion ratio areas, and the number of points per peak be investigated more criterion for each of the target PPCPs was applied to real carefully in the confirmation of these compounds. environmental samples that were not spiked with standards. Once the acceptable tolerance range for the mean ion *ese samples were wastewater influents and effluents and ratio for each analyte was established based on spiked en- receiving surface waters which are located upstream and vironmental samples, the ion ratio in each sample matrix downstream of the respective WWTPs, collected from 5 was also assessed in order to adjust this range accordingly for different countries. For influent samples, an analyte is said to the influent, effluent, upstream, and downstream samples. be positively detected in the sample if its ion ratio is within *e tolerance range that would give ≤5% false negatives was the mean ± 40% of the reference standard. For effluent, recorded for each matrix type. *ese values were 40% for the upstream, and downstream samples, analytes with the mean influent samples and 30% for the effluent, upstream, and ion ratio within ±30% of the standards are considered downstream samples (Table 3). It is expected that the positive detection. Note that the results in Table 6 do not compounds would have higher variations in more complex include detections for tetracyclines and quinolones, for matrices such as the influent. Since the tolerance range for which a different tolerance range was set. the influent differed by 10%, it is recommended to establish *e compounds that were detected outside the range a different tolerance limit for influent samples to avoid were acetylsulfamethoxazole in the effluent samples, azi- a high false-negative rate in this matrix. If a fixed range of thromycin in the effluent and upstream samples, and clar- ±20% is used as the tolerance value (Table 2) for all types of ithromycin in the downstream samples, with the details matrices, the number of false negatives would be much shown in Table 7. 8 Journal of Analytical Methods in Chemistry

Table 7: Compounds with ion ratios detected outside the set 100 tolerance range of ±40% for WWTP influent samples and 30% for WWTP effluent, upstream, and downstream samples. 90 Tolerance Calculated 80 Matrix Compound range ion ratio 70 Acetylsulfamethoxazole 0.88–1.64 1.74 Effluent Azithromycin 1.89–3-5 1.86 60 Clarithromycin 1.28–2.38 2.41 Downstream Clarithromycin 1.28–2.38 2.42 50 Upstream Azithromycin 1.89–3-5 1.74 40 Relative abundance Relative *e compounds with ion ratios that fell outside the set 30 tolerance range were investigated individually to confirm if these were real detections or not by checking the presence of 20 both the quantitative and qualitative ions and if the shift in 10 retention time is not more than 0.5 min. It was found that all of them had both ions, and their retention times were within 0 the acceptable range. Since their calculated ion ratios are still 20 22 24 26 28 30 close to the limits of the range, these were still considered as (a) positive detections. In cases like this where the calculated ion 100 ratios are close to the limits of the range and retention times are within the acceptable shift, it is recommended that the 90 qualitative ion be checked to make sure that its signal is at 80 least 3 times that of the noise (Table 6). A total number of 37 detections for ciprofloxacin, 70 norfloxacin, and tetracycline were found for tetracyclines and quinolones in the samples. *e ion ratios of all 37 peaks 60 in all matrices were within the set tolerance range of 80%, 50 and they passed other criteria for peak confirmation. An example of a false-positive detection that was found 40

through the use of the ion ratio is diclofenac. *e quantitative abundance Relative and qualitative ion transitions of diclofenac are 296 → 214 and 30 296 → 250 and its retention time is at 27.5 min. Figure 5 shows 20 a comparison of the two chromatograms, both of which have peaks at 27 min for both SRM transitions. 10 When the ion ratios were calculated, an influent sample (Figure 5) gave an ion ratio of 0.15, which falls outside the 0 range for diclofenac in the influent which is from 1.57 to 20 22 24 26 28 30 3.67. Furthermore, it can be observed that, for WWTP A, the (b) retention time of the qualitative ion, which is at 27.16 min, is slightly different from that of the quantitative one at Figure 5: False-positive detection of diclofenac in wastewater. *e 27.78 min, further proving that this is a false-positive de- calculated ion ratio, 0.15, falls outside the ion ratio tolerance range tection since the retention times of both ions should be the of 1.57 to 3.67. *e chromatograms show (a) the peak for the quantitative ion with a transition of 296 → 214 and (b) the peak for same. Upon removal of 7 false-positive diclofenac peaks that the qualitative ion with a transition of 296 → 250. did not match the ion ratio criterion and retention times for both the quantitative and qualitative ions, the total number of detections was reduced from 312 to 305. ion ratios from 1.56, 3.13, 6.25, 12.5, 25, 50, and 100 ppb so that it can capture the variations at different concentrations. 4. Conclusions *e ion ratios for tetracyclines and quinolones were found to have higher variations, which are twice that of the other An ion ratio criterion has been optimized for six classes of PPCPs; therefore, these two classes of compounds were pharmaceuticals in wastewater and surface water using LC- analyzed separately so as not to increase the possibility of MS/MS. For 23 PPCPs, values for mean ± tolerance for the false positives for the other compound classes. For tetra- ion ratios in the different types of environmental matrices cyclines and quinolones, the tolerance range was set to 80%, but were established based on the variabilities of the ion ratios in it is recommended that other criteria such as retention time, spiked samples. *e variabilities of the ion ratios of the peak areas, and the number of points per peak be investigated compounds were found to increase at lower concentrations carefully before reporting their detections. from 4% at 100 ppb to 18% at 1.56 ppb. *erefore, the mean For the sulfonamides, macrolides, SSRIs, and other ion ratio that was used in the formula is the average of the PPCPs, the ion ratios were studied in the different Journal of Analytical Methods in Chemistry 9 environmental matrices. It was found that the variations also [8] L. Rizzo, C. Manaia, C. Merlin et al., “Urban wastewater increase with the complexity of the matrix. *e optimized treatment plants as hotspots for antibiotic resistant bacteria tolerance range that would give <5% false negatives was 40% and genes spread into the environment: a review,” Science of for the influent and 30% for the effluent, upstream, and the Total Environment, vol. 447, pp. 345–360, 2013. downstream. *is optimized ion ratio criterion was then [9] K. Kümmererand A. Henninger, “Promoting resistance by the emission of antibiotics from hospitals and households into applied to real wastewater and surface water samples that effluent,” Clinical Microbiology and Infection, vol. 9, no. 12, were not spiked with standards and resulted in the reduction pp. 1203–1214, 2003. of the total number of detections from 312 to 305, after false [10] N. H. Tran, M. Reinhard, and K. Y. H. Gin, “Occurrence and fate positives were eliminated. of emerging contaminants in municipal wastewater treatment plants from different geographical regions-a review,” Water Data Availability Research, vol. 133, pp. 182–207, 2018. [11] WHO, Antimicrobial Resistance, WHO, Geneva, Switzerland, All data underlying the findings of this study can be accessed 2017, http://www.who.int/mediacentre/factsheets/fs194/en/. in the supplementary information provided. [12] H. L. Schoenfuss, E. T. Furlong, P. J. Phillips et al., “Complex mixtures, complex responses: assessing pharmaceutical mixtures using field and laboratory approaches,” Environmental Conflicts of Interest Toxicology and Chemistry, vol. 35, no. 4, pp. 953–965, 2016. [13] N. H. Tran, H. Chen, M. Reinhard, F. Mao, and K. Y. Gin, *e authors declare that they have no conflicts of interest. “Occurrence and removal of multiple classes of antibiotics and antimicrobial agents in biological wastewater treatment Acknowledgments processes,” Water Research, vol. 104, pp. 461–472, 2016. [14] B. J. Berendsen, T. Meijer, H. G. Mol, L. van Ginkel, and *e authors would like to acknowledge support from the Na- M. W. Nielen, “A global inter-laboratory study to assess tional Science Foundation (PIRE-HEARD, Award no. 1545756). acquisition modes for multi-compound confirmatory analysis of veterinary drugs using liquid chromatography coupled to Supplementary Materials triple quadrupole, time of flight and orbitrap mass spectrometry,” Analytica Chimica Acta, vol. 962, pp. 60–72, 2017. Table S1: retention time and transitions of pharmaceuticals [15] FDA, Guidance for Industry 118 Confirmation of Identity of in the study. Table S2: average ion ratios of pharmaceuticals Animal Drug Residues, Food and Drug Administration, Silver at different concentrations. (Supplementary Materials) Spring, MD, USA, 2003. [16] UNODC, Guidance for the Validation of Analytical Meth- odology and Calibration of Equipment Used for Testing of Illicit References Drugs in Seized Materials and Biological Specimens, UNODC, New York, NY, USA, 2009. [1] P. Arnnok, R. R. Singh, R. Burakham, A. Pérez-Fuentetaja, [17] *e European Communities, Commission Decision 2002/657/ and D. S. Aga, “Selective uptsake and bioaccumulation of EC Implementing Council Directive 96/23/EC Concerning the antidepressants in fish from effluent-impacted Niagara River,” Performance of Analytical Methods and the Interpretation of Environmental Science and Technology, vol. 51, no. 18, Results, *e European Communities, 2002. pp. 10652–10662, 2017. [18] USDA, Data and Instrumentation Revision 5. Agricultural [2] K. Kummerer, “Antibiotics in the aquatic environment–a Marketing Service, United States Department of Agriculture, review–part I,” Chemosphere, vol. 75, no. 4, pp. 417–434, 2009. Washington, DC, USA, 2017. [3] I. Senta, I. Krizman-Matasic, S. Terzic, and M. Ahel, “Com- [19] EWDTS, European Laboratory Guidelines for Legally Defensible prehensive determination of macrolide antibiotics, their syn- Workplace Drug Testing (EWDTS), CRC Press, BocaRaton, FL, thesis intermediates and transformation products in wastewater USA, 2002, http://www.eapinstitute.com/documents/EWDTS- effluents and ambient waters by liquid chromatography-tandem Guidelines.pdf. mass spectrometry,” Journal of Chromatography A, vol. 1509, [20] H. G. Mol, P. Zomer, M. Garcia Lopez et al., “Identification in pp. 60–68, 2017. residue analysis based on liquid chromatography with tandem ´ [4] I. Senta, S. Terzic, and M. Ahel, “Simultaneous determination of mass spectrometry: experimental evidence to update perfor- sulfonamides, fluoroquinolones, macrolides and trimethoprim mance criteria,” Analytica Chimica Acta, vol. 873, pp. 1–13, 2015. in wastewater and river water by LC-tandem-MS,” Chroma- [21] B. J. Berendsen, T. Meijer, R. Wegh et al., “A critical as- tographia, vol. 68, no. 9-10, pp. 747–758, 2008. sessment of the performance criteria in confirmatory analysis [5] L. J. Zhou, G. G. Ying, S. Liu et al., “Occurrence and fate of for veterinary drug residue analysis using mass spectrometric eleven classes of antibiotics in two typical wastewater treat- detection in selected reaction monitoring mode,” Drug Testing ment plants in South China,” Science of the Total Environ- and Analysis, vol. 8, no. 5-6, pp. 477–490, 2016. ment, vol. 452-453, pp. 365–376, 2013. [22] USEPA, Method 1694: Pharmaceuticals and Personal Care [6] M. Pedrouzo, F. Borrull, R. M. Marce, and E. Pocurull, “Ultra- Products in Water, Soil, Sediment, and Biosolids by HPLC/MS/ high-performance liquid chromatography-tandem mass spec- MS, U.S Envorinmrntal Protection Agency, Washington, DC, trometry for determining the presence of eleven personal care USA, 2007. products in surface and wastewaters,” Journal of Chromatog- [23] USEPA, Method 542: Determination of Pharmaceuticals and raphy A, vol. 1216, no. 42, pp. 6994–7000, 2009. Personal Care Products in Drinking Water by Solid Phase Ex- [7] E. Kristiansson, J. Fick, A. Janzon et al., “Pyrosequencing of traction and Liquid Chromatography Electrospray Ionization– antibiotic-contaminated river sediments reveals high levels of Tandem Mass Spectrometry (LC/ESI-MS/MS), U.S Envorinmrntal resistance and gene transfer elements,” PLoS one, vol. 6, no. 2, Protection Agency, Washington, DC, USA, 2016. article e17038, 2011. Hindawi Journal of Analytical Methods in Chemistry Volume 2018, Article ID 7965124, 8 pages https://doi.org/10.1155/2018/7965124

Research Article

Determination of Tobramycin in M9 Medium by LC-MS/MS: Signal Enhancement by Trichloroacetic Acid

Liusheng Huang ,1 Janus Anders Juul Haagensen,2 Davide Verotta ,1 Vincent Cheah,1 Alfred M. Spormann,3 Francesca Aweeka,1 and Katherine Yang 1

1Department of Clinical Pharmacy, School of Pharmacy, University of California San Francisco, San Francisco, CA, USA 2Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark, 2800 Kongens Lyngby, Denmark 3Department of Civil and Environmental Engineering, Stanford University, Palo Alto, CA, USA

Correspondence should be addressed to Liusheng Huang; [email protected] and Katherine Yang; [email protected]

Received 8 December 2017; Accepted 6 February 2018; Published 26 April 2018

Academic Editor: Lucia Mendez

Copyright © 2018 Liusheng Huang et al. ,is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

It is well known that ion-pairing reagents cause ion suppression in LC-MS/MS methods. Here, we report that trichloroacetic acid increases the MS signal of tobramycin. To support studies of an in vitro pharmacokinetic/pharmacodynamic simulator for bacterial biofilms, an LC-MS/MS method for determination of tobramycin in M9 media was developed. Aliquots of 25 μLM9 media samples were mixed with the internal standard (IS) tobramycin-d5 (5 µg/mL, 25 µL) and 200 µL 2.5% trichloroacetic acid. ,e mixture (5 µL) was directly injected onto a PFP column (2.0 × 50 mm, 3 µm) eluted with water containing 20 mM ammonium formate and 0.14% trifluoroacetic acid and acetonitrile containing 0.1% trifluoroacetic acid in a gradient mode. ESI+ and MRM with ion m/z 468 → 324 for tobramycin and m/z 473 → 327 for the IS were used for quantification. ,e calibration curve concentration range was 50–25000 ng/mL. Matrix effect from M9 media was not significant when compared with injection solvents, but signal enhancement by trichloroacetic acid was significant (∼3 fold). ,e method is simple, fast, and reliable. Using the method, the in vitro PK/PD model was tested with one bolus dose of tobramycin.

1. Introduction to its higher hydrophilicity and (2) lack of chromophores for detection. Numerous assays have been reported including Tobramycin (TBM) is an aminoglycoside antibiotic widely HPLC coupled with UV [3], electrochemical [4, 5], or ﬂuo- used for the treatment of multidrug-resistant Gram-negative rescence detectors [6], and these assays lack sensitivity and bacterial infections by inhibiting protein synthesis and al- usually require derivatization. LC-MS/MS assays have also tering integrity of the bacterial cell membrane [1]. It is also been reported, but the sensitivity of these assays requires named 3′-deoxykanamycin B, nebramycin 6, and chemically concentrations ≥100 ng/mL [7–9]. Trichloroacetic acid (TCA) O-3-amino-3-deoxy-α-D-glucopyranosyl-(1-6)-O-[2,6-dia- has been used in sample preparation to remove proteins, es- mino-2,3,6-trideoxy-α-D-ribo-hexopyranisyl-(1-4)]-2-deoxy- pecially for hydrophilic analytes, with the advantage of direct D-streptamine (Figure 1). It is water soluble and stable at injection of resulting sample solution [10]. We found that TCA ambient temperature at a wide range of pH 1–11 [2]. not only increased the retention time but also the MS signal of To support pharmacokinetic (PK) and pharmacodynamic TBM. Built on this observation, we report a simple LC-MS/MS (PD) studies of TBM for bioﬁlm-mediated infections using an method to determine TBM in M9 medium using TCA as the in vitro model, an analytical method to quantitate TBM in M9 ion pair reagent in the injection sample instead of the mobile medium is needed. Two considerable challenges in de- phase. In addition, this assay utilized a PFP column, which termination of TBM in biological matrices are (1) poor re- yielded a better retention factor for TBM (k � 1.8). ,e cali- tention on commonly used reverse-phase HPLC columns due bration range was 50–25000 ng/mL. 2 Journal of Analytical Methods in Chemistry

H N O 2 OH HO O H N NH 2 HO 2 OH O O

NH2 H2N 163 324

(a) 3000,000 468 2000,000 163 1000,000 324 Intensity (cps) Intensity 0 0 50 100 150 200 250 300 350 400 450 500 m/z 3000,000 473 2000,000

164 1000,000 327 Intensity (cps) Intensity 0 0 50 100 150 200 250 300 350 400 450 500 m/z

(b)

Figure 1: Product ion spectra of tobramycin (a) and deuterated tobramycin (b).

2. Experimental a gradient program consisting of 5% solvent B (0–0.10 min), from 5 to 20% B (0.10–1.50 min), from 20 to 80% B (1.50– 2.1. Chemicals and Reagents. Tobramycin was purchased 1.51 min), 80% B (1.51–2.00 min), 80%–5% B (2.00–2.01 min), from Sigma-Aldrich (St. Louis, MO, USA). Deuterated and 5% B (2.01–3.00 min). Retention times for TBM and the tobramycin (TBM-d5) was purchased from TorontoResearch internal standard (IS) were both 0.84 min. ,e divert valve was Chemicals (North York, Ontario, Canada). Formulated set to direct the LC eluent to the mass spectrometer (MS) tobramycin (20 mg/2 mL, for IM or IV use) was obtained source at 0.6 min and to the waste line at 2.9 min. ,e MS from APP Pharmaceuticals, LLC (Schaumburg, IL, USA). conditions for TBM and the IS were optimized by separate Common solvents (HPLC grade) and reagents (Certified infusion of 200 ng/mL TBM and 400 ng/mL deuterated TBM ACS) were obtained from ,ermo-Fisher Sci. (Fair Lawn, NJ, in 0.1% formic acid into the MS at a flow rate of 15 and USA). M9 minimal salts ×5 solution was prepared by dis- 25 µL/min constantly while adjusting MS parameters with solving 2.82 g Difco™ M9 minimal salts (BD, Sparks, MD, autotune followed by manual adjustment to achieve the USA) in 50 mL water. M9 medium was prepared by adding maximal signal. ,e ions m/z 468 → 324 for TBM and m/z 10 mL M9 minimal salts ×5 solution, 5 µL 1 M CaCl2, 50 µL 473 → 327 for the IS were used for quantification in the 1 M MgSO4, and 13.5 µL 20% glucose to 40 mL water. multiple reaction monitoring (MRM) mode. ,e optimized compound-dependent MS parameters were 121 V (DP), 21 V 2.2. Instrumental. ,e LC-MS/MS system consists of an AB (CE), and 26 V (CXP) for both TBM and the IS. DP was Sciex API5000 Tandem Mass Spectrometer, two Shimadzu declustering potential, CE was collision energy, and CXP was Prominence 20ADXR UFLC pumps, and an SIL-20ACXR collision cell exit potential. ,e instrument-dependent pa- autosampler managed with Analyst® 1.6.2 (AB Sciex, Redwood rameters were optimized by flow injection analysis (FIA): an City, CA, USA). ,e gases for the MS system were supplied by an aliquot of 5 µL 200 ng/mL TBM was repeatedly injected into LC-MS gas generator (Source 5000™, Parker Balston Inc., the LC-MS/MS system while LC flow was maintained at Haverhill, MA, USA). LC conditions were as follows: separation 0.4 mL/min 50% B isocratically without column in the line. was achieved on a Pursuit PFP column (2.1 × 50 mm, 3 µm) ,e optimized MS parameters were as follows: MS source was (Agilent Tech.Inc., Santa Clara, CA, USA). Mobile phase A was the TurboIon Spray ionization in positive mode (ESI+) with ° 20 mM NH4FA 0.14% trifluoroacetic acid (TFA) and B was turbo heater set at 500 C, curtain gas was nitrogen at 40 psi, 0.1% TFA in acetonitrile (MeCN). Five-microliter sample was nebulizer gas (gas 1) and auxiliary (Turbo)gas (gas 2) were zero injected onto the column eluted at a flow rate of 0.4 mL/min in air set at 50 psi and 60 psi, respectively, collision-deactivated Journal of Analytical Methods in Chemistry 3 association gas was nitrogen at 12 psi, and ionspray voltage was −70°C. Stability of freshly prepared IS working solution was 5500 V. Data were processed with Analyst 1.6.2. (AB Sciex, evaluated at room temperature for 24 hr and 5 days. Effects Redwood City, CA, USA). of concomitant drugs (e.g., meropenem and colistin) on quantification were evaluated by spiking them in the QC samples at a final concentration of 110 µg/mL meropenem 2.3. Preparation of Calibrators, Quality Controls, and Internal (MP) and 20 µg/mL colistin. ,e measured concentrations of Standard. As TBM used in the in vitro biofilm PK/PD model TBM were compared to the QC samples without these contains formulation ingredients, calibrators and quality concomitant drugs. controls (QCs) were prepared from formulated TBM (20 mg/2 mL) with serial dilution in M9 medium to match the matrix in unknown samples. Calibrators consists of 50, 100, 250, 2.6. Application. ,is method was used to validate a novel 500, 1000, 2500, 5000, 10000, and 25000 ng/mL. QCs consist of dynamic PK/PD model designed to study the effects of human- 150, 1500, 20000, and 40000 ng/mL, designated as low-, me- simulated antibiotic concentrations on Pseudomonas aerugi- dium-, high-, and extrahigh QC. ,e internal standard TBM-d5 nosa biofilms grown in vitro [11]. TBM, in conjunction with solution was prepared in water by serial dilution at a final a β-lactam antibiotic such as MP, is recommended for the concentration of 5000 ng/mL. ,e IS solution needs to stand on treatment of multidrug-resistant Pseudomonas aeruginosa bench overnight before use. lung infection in patients with cystic fibrosis [12]. While the formation of bacterial biofilms in the lung is a characteristic of chronic lung infection in patients with cystic fibrosis, the PD of 2.4. Sample Preparation. M samples (25 µL) were pipetted 9 antibiotics on biofilms is largely unknown. ,e concentration- into 1 mL glass autosampler vials, to which were added 25 µL time curves of single and multiple intravenous bolus doses of IS (5 µg/mL TBM-d ) and 200 µL 2.5% TCA. After vortex 5 TBM were simulated based on human population PK pa- mixed, the samples were placed in the autosampler tray. If rameters [13]. ,e target TBM peak concentration, based on the samples were collected from M medium flowing 9 a dose of 10 mg/kg in a 70 kg adult, was 32.79 mg/L with an through bacterial biofilm, the samples were centrifuged at associated t � 2.75 h. Samples were taken at t � 0, 1, 2, 4, 6, 8, 20000g for 3 min before adding to the sample vial. Injection 1/2 16, and 24 hr from the main feeding bottle and the tubing volume was 5 µL. outlets from three flow cells with bacterial biofilm. All samples were shipped to our analytical lab on the same day with dry ice 2.5. Validation. ,e method was validated in terms of overnight delivery and stored at −70°C freezer until analysis. precision, accuracy, matrix effect, and stability, following the Samples were typically analyzed within a week. procedures as described previously [10]. One set of calibrators was processed for each run and injected in the 3. Results and Discussion beginning of the batch run. Calibration curves were constructed by linear regression of the peak area ratio of the 3.1. LC-MS/MS Optimization. TBM contains five amine analyte to the IS (y-axis) versus the nominal analyte con- groups (Figure 1), making electrospray ionization in positive centrations (x-axis) with a weighting factor of 1/x. ,e lower mode (ESI+) the choice of the ion source. ,e ion m/z limit of quantification (LLOQ) was established with pre- 468 → 324 was chosen for quantification for its signal cision and accuracy <20%. Intraday precision and accuracy abundancy and selectivity. Compared to product ion m/z 163, were determined by analysis of at least five replicates of each m/z 324 has less background signal. ,e deuterated TBM was QC sample at low (150 ng/mL), medium (1500 ng/mL), and used as the IS. However, the deuteration positions were not high (20000 ng/mL) concentration levels extracted with a set identified. MS scan showed that multiple forms of deuterated of calibrators in one batch. ,e same procedure was repeated TBM exist, with the most abundant protonated molecule at on at least 2 different days with new samples to determine m/z 473. ,erefore, ion m/z 473 → 327 was chosen for the IS. interday precision and accuracy (total: n ≥ 15 per concen- ,e signal of the ion m/z 473 → 327 decreased gradually in the tration level). Precision was reported as relative standard first few hours but remained stable after the IS solution stood deviation (RSD) and accuracy as percent deviation from the on bench overnight. ,ese observations suggested that nominal concentration (% dev.). Matrix effect was evaluated deuteration most likely occurred on amine groups, and the as follows: TBM was spiked at the concentrations of 300, stable form of IS contains a deuterium atom on each amine 1500, and 20000 ng/mL in water and M9 medium, re- group (Figure 1). spectively. ,ree aliquots of each sample were processed as Having 5 amine groups and 5 hydroxyl groups also described above (Section 2.4). ,e peak areas and peak area makes TBM hardly retain on reverse-phase LC columns. Ion ratios of TBM in M9 medium were compared to those in pair reagent TFA and TCA in the mobile phase could help to water. Values within 100 ± 15% were considered as no sig- retain polar amino molecules on the reverse-phase columns; nificant matrix effect from M9 medium. To evaluate partial however, sensitivity may be compromised due to ion sup- volume accuracy, 12.5 µL extrahigh QC at 40000 ng/mL was pression. Previously, we found that TFA could change re- mixed with 12.5 µLM9 medium and processed as described tention time of isoniazid when added into sample before in Section 2.4. Stability was evaluated in the following injection (Supplementary Material Figure S1). However, conditions: room temperature (21–25°C) for 5 days, 3 days TFA did not improve the TBM peak. Cheng et al. used TCA to on the autosampler rack, 3 freeze-thaw cycles, and 6 days at modify retention time of aminoglycoside compounds [14]. 4 Journal of Analytical Methods in Chemistry

3000 25,000

2500 TBM 20,000 2000 TBM–d5 15,000 1500 10,000 1000 Intensity (cps) Intensity Intensity (cps) Intensity 500 5000

0 0 0123 0 123 Time (min) Time (min)

(a) (b) Figure 2: Chromatograms of blank M9 medium (blue solid line), blank M9 medium spiked with IS (red dash line), and TBM at LLOQ level (black solid line). (a) TBM channel; (b) the IS channel.

14,000 Mobile phase: A = 10 mM NH4FA, B = 0.1% FA in acetonitrile 12,000

10,000

8000

6000

Intensity (cps) Intensity 4000

2000

0 0 0.5 1 1.5 2 2.5 3 Time (min)

Water 2% TCA

(a) 14,000 Mobile phase: A = 20 mM NH4FA 0.14% TFA, B = 0.1% TFA in acetonitrile 12,000 10,000 8000 6000 4000 Intensity (cps) Intensity 2000 0 0 0.5 1 1.5 2 2.5 3 Time (min)

Water 2% TCA

(b)

Figure 3: Impact of mobile phase solvents and sample solvents on peak shape, retention time, and signal intensity of TBM. Sample solvents: water (dash line) and 2% TCA (solid line). Mobile phase solvents: 10 mM NH4FA (pH 4.0)-0.1% FA in MeCN (a) and 20 mM NH4FA 0.14% TFA-0.1% TFA in MeCN (b).

We found that when the sample contained 2% TCA with retention time tR � 0.839 min, the estimated dead volume is a 5 µL injection volume, longer retention time of TBM 0.68πr2L � 0.118 mL, and retention factor k � 1.84. was observed (Supplementary Material Figure S2). Under Unexpectedly, TCA also enhanced MS response of TBM. the final LC condition, the TBM peak was sufficiently sep- Two different sample solvents (water and 2% TCA) and two arated from the matrix-generated peaks (Figure 2). ,e sets of mobile phase solvents were tested: (1) A � 10 mM Journal of Analytical Methods in Chemistry 5

Table 1: Interday average backcalculated standard concentrations (n � 3). Nominal concentration (ng/mL) 50 100 250 500 1000 2500 5000 10000 25000 R Mean (ng/mL) 50.1 91.6 236 521 1043 2663 4973 9977 24800 0.9992 Precision (RSD, %) 3.82 6.11 3.31 2.98 3.37 2.50 8.06 3.72 2.91 0.0379 Accuracy (% dev.) 0.13 −8.37 −5.60 4.13 4.33 6.53 −0.53 −0.23 −0.80 n 3 3 3 3 3 3 3 3 3

Table 2: Intra- and interday precision and accuracy. Intraday Interday Nominal (ng/mL) 50.0 150 1500 20000 50.0 150 1500 20000 Mean (ng/mL) 43.0 to 50.9 150 to 159 1533 to 1653 20450 to 20650 46.8 153 1591 20572 Precision (RSD) (%) 3.0 to 16.9 4.4 to 6.7 2.1 to 3.5 2.5 to 3.4 8.43 3.33 3.78 0.52 Accuracy (dev.) (%) −14.0 to 1.7 0 to 5.9 2.2 to 10.2 2.3 to 3.3 −6.44 1.96 6.04 2.86 n 6 6 6 6 3 3 3 3

Table 3: Matrix eﬀect. TBM peak area (×104) IS peak area (×104) Ratio Matrix eﬀect Concentration (ng/ml) Water M9 Water M9 Water M9 TBM IS Ratio Low (120) 3.60 ± 0.12 3.83 ± 0.13 8.26 ± 0.49 8.72 ± 0.53 0.436 0.439 106 106 101 Medium (1500) 18.1 ± 1.0 20.0 ± 1.4 8.80 ± 0.77 9.30 ± 0.91 2.06 2.15 110 106 104 High (17000) 378 ± 5 413 ± 18 13.7 ± 0.5 14.7 ± 0.7 27.6 28.1 109 107 102 Data represent the mean peak area (±SD) from triplicate analysis.

NH4FA at pH 4.0; B � 0.1% FA in MeCN and (2) A � 20 mM intraday/interday precision and accuracy and interference of NH4FA 1.4% TFA;, B � 0.1% TFA in MeCN, using the same concomitant drugs were repeated with the new low QC gradient elution method. With the commonly used mobile concentration. phase solvents (set 1), the peak shape for TBM was poor if injection solvent is water, while 2% TCA in the sample im- 3.2.1. Calibration Range. At the LLOQ concentration proved peak shape, signal intensity, and retention time sig- (50 ng/mL), the signal intensity was 2100–2400 cps (peak nificantly (Figure 3(a)); with mobile phase solvent set 2, the area, 6600–7900) and signal-to-noise ratio S/N � 30–48 signal intensity and retention time of TBM improved further (Figure 2). ,is LLOQ is lower than others reported in lit- (Figure 3(b)). ,is improvement is critical as the interference erature. A recent study reported an LLOQ at 100 ng/mL. ,e peak from M9 medium was then separated from the TBM detector was the same as ours, but heptafluorobutyric acid peak (Figure 2). ,e exact mechanism of signal enhancement was used as ion pair reagent in the mobile phase and sample by TCA is unknown. Cheng et al. thought that reduced matrix reconstitution [9]. ,e calibration curve was constructed with effect with longer retention time contributed to the signal least square linear regression weighted by 1/x. ,e interday enhancement [14], but we observed signal enhancement in backcalculated concentrations of calibrators over 3 days are neat solution (Figure 3). ,e possible reason could be that listed in Table 1. ,e precision is within 10% and accuracy TCA limited multiple charges of TBM and thus increased (percent deviation from the nominal value) is within ±10%, + monocharged molecular ion ([M+H] ). In addition, we too. Representative MRM ion chromatograms of TBM from observed that MS response of the IS (TBM-d5) was also in- M9 medium (double blank), M9 medium spiked with IS creased with the increase of TBM concentration, suggesting (blank), and LLOQ samples are shown in Figure 2. ion enhancement of coeluting compounds. ,is should not affect quantification as IS was added to all samples, and the TBM signal increased accordingly. ,is was confirmed with 3.2.2. Precision and Accuracy. ,e intraday precision (n � 6) the excellent linearity of calibration curve. was within 7% at low, medium, and high concentrations. ,e interday precision, calculated with the individual mean concentration from 3 days, was within 5% at the three 3.2. Validation. Based on our initial simulation, the TBM concentration levels. ,e intra- and interday accuracy was all trough concentration is expected to be >250 ng/mL. ,erefore, within 15%. At the LLOQ levels, the precision and accuracy the LLOQ in this assay was initially set at 250 ng/mL, the upper met the criteria of <20% (Table 2). limit of quantification was set at 25000 ng/mL, and validation was performed with low (300 ng/mL), medium (1500 ng/mL), and high (20000 ng/mL) QCs. After tested the in vitro biofilm 3.2.3. Matrix Effect. ,e matrix effect of M9 medium on model, we found that the trough TBM concentration fell below both TBM and IS signals is within 100 ± 15% (Table 3), 250 ng/mL, and thus, we lowered the LLOQ to 50 ng/mL and suggesting that the matrix effect of M9 medium was not the low QC level to 150 ng/mL accordingly. Validation of significant. ,e matrix effect on the peak area ratio was 6 Journal of Analytical Methods in Chemistry

Table 4: Stability of TBM. Conditions % remained RSD (%) n In autosampler vial, 21–25°C, 3 days 300 ng/mL 107 3.5 3 20000 ng/mL 105 2.4 3 ° In M9, 21–25 C, 5 days 300 ng/mL 104 8.1 3 20000 ng/mL 99.2 3.6 3 3 freeze-thaw cycles 300 ng/mL 99.4 5.0 3 20000 ng/mL 99.8 2.4 3 ° In M9 medium, 6 days, −70 C 300 ng/mL 93.8 3.3 3 20000 ng/mL 101 3.0 3 IS (5000 ng/mL) in water 102.9 0.74 3 24 hr, 21–25°C 74.2 3.4 3 5 days, 21–25°C 79.5 1.7 4

Table 5: Interference of potential concomitant drugs. Concentration (ng/mL) Control Colistin-MP∗ % dev. 150 150 ± 10 148 ± 8 −1.3 1500 1633 ± 61 1587 ± 98 −2.8 20000 20000 ± 557 19933 ± 737 −0.3 Note. Data represent the mean (SD) of triplicate analysis. ∗MP and colistin concentrations were 110 µg/mL and 20 µg/mL, respectively, corresponding to the highest concentrations in the in vitro model.

100,000

10,000

1000

100

TBM concentration (ng/mL) TBM concentration 10

1 0 5 10 15 20 25 Time (hr)

M F2 F1 F3

Figure 4: Concentration-time profile of tobramycin from an in vitro PK/PD biofilm model. Samples were taken from the feeding bottle (M) and the tubing outlets from three flow cells with bacterial biofilm (F1, F2, and F3) at designated time intervals. even smaller, suggesting that IS compensated the matrix analysis were 2.2% and 2.6%, respectively. ,erefore, sam- effect. ples above the upper limit of quantification could be analyzed with a partial volume.

3.2.4. Partial Sample Volume Accuracy. As the target Cmax 3.2.5. Stability. TBM was stable in M9 medium. No sig- is 40000 ng/mL, we evaluated accuracy of the assay with an niﬁcant degradation was found under tested condition extrahigh QC (40000 ng/mL) when half sample volume was (Table 4). Further investigation is ongoing to deﬁne long- used. ,e precision and accuracy from six replicates of term stability in −70°C freezer. Journal of Analytical Methods in Chemistry 7

3.2.6. Evaluation of Concomitant Drug Interference. ,e Department of Clinical Pharmacy at the University of samples from the supported study are expected to contain California San Francisco. MP and colistin; therefore, impact of these drugs on quantification of TBM was evaluated. In the presence of 110 µg/mL MP and 20 µg/mL colistin, the low, medium, and Supplementary Materials high QC samples could still be quantified accurately, with Figure S1: effect of TFA on retention time of isoniazid. a small percent deviation from the samples without these Figure S2: optimization of TCA concentration. (Supple- drugs (Table 5). mentary Materials)

3.3.Application. ,e method was applied to determine TBM References concentrations used in an in vitro PK/PD biofilm simulator. ,e PK/PD analysis was reported elsewhere [15]. A repre- [1] E. J. Begg and M. L. Barclay, “Aminoglycosides—50 years on,” sentative concentration-time curve from the model is British Journal of Clinical Pharmacology, vol. 39, no. 6, pp. 597–603, 1995. showed in Figure 4. ,e results demonstrate that the sen- [2] H. C. Neu, “Tobramycin: an overview,” Journal of Infectious sitivity of the method met the requirement of the intended Diseases, vol. 134, pp. S3–S19, 1976. study. [3] C. H. Feng, S. J. Lin, H. L. Wu, and S. H. Chen, “Trace analysis of tobramycin in human plasma by derivatization and high- 4. Conclusion performance liquid chromatography with ultraviolet detection,” Journal of Chromatography B, vol. 780, no. 2, TCA not only improves peak shape and retention time of pp. 349–354, 2002. TBM but also increases MS signal intensity of TBM. Using [4] J. A. Statler, “Determination of tobramycin using high- a simple dilution with ion pairing reagent TCA, a sensitive performance liquid chromatography with pulsed ampero- LC-MS/MS method was developed and validated for de- metric detection,” Journal of Chromatography B: Biomedical Sciences and Applications, vol. 527, pp. 244–246, 1990. termination of TBM in bacterial M9 medium. ,e LLOQ was [5] C. Ghinami, V. Giuliani, A. Menarini, F. Abballe, S. Travaini, 50 ng/mL. ,e sensitivity of the assay met the requirement of and T. Ladisa, “Electrochemical detection of tobramycin or the intended PK/PD study in an in vitro biofilm model gentamicin according to the European Pharmacopoeia ana- system. lytical method,” Journal of Chromatography A, vol. 1139, no. 1, TCA has been used to increase retention time and pp. 53–56, 2007. sensitivity for quantification of gentamicin, kanamycin, and [6] F. Lai and T. Sheehan, “Enhancement of detection sensitivity apramycin [14]. Here, we demonstrated application of TCA and cleanup selectivity for tobramycin through pre-column to quantification of TBM. We speculate this approach could derivatization,” Journal of Chromatography A, vol. 609, no. 1- be generalized: by addition of ion-pairing agents to samples 2, pp. 173–179, 1992. instead of adding to mobile phase solvents, we could extend [7] B. G. Keevil, S. J. Lockhart, and D. P. Cooper, “Determination of tobramycin in serum using liquid chromatography-tandem the retention time of analytes and even increase sensitivity. mass spectrometry and comparison with a fluorescence Acidic ion-pairing agents such as TFA and TCA could be polarisation assay,” Journal of Chromatography B, vol. 794, applied to basic polar analytes such as amine-containing no. 2, pp. 329–335, 2003. analytes, and basic ion-pairing agents could be added to [8] M. X. Guo, L. Wrisley, and E. Maygoo, “Measurement of samples of acidic polar analytes. Nevertheless, the concen- tobramycin by reversed-phase high-performance liquid tration of the ion-pairing agent is critical, and selection of chromatography with mass spectrometry detection,” Analy- the ion-pairing agent is also critical. tica Chimica Acta, vol. 571, no. 1, pp. 12–16, 2006. [9] P. M. Bernardi, F. Barreto, and T. Dalla Costa, “Application of a LC-MS/MS method for evaluating lung penetration of Disclosure tobramycin in rats by microdialysis,” Journal of Pharma- ceutical and Biomedical Analysis, vol. 134, pp. 340–345, 2016. ,is work was presented in the ACS National Meeting (http:// [10] L. Huang, P. Lizak, C. C. Dvorak, F. Aweeka, and J. Long- sanfrancisco2017.acs.org/i/803418-253rd-american-chemical- Boyle, “Simultaneous determination of fludarabine and clo- society-national-meeting-expo/89). Its contents are solely farabine in human plasma by LC-MS/MS,” Journal of Chro- the responsibility of the authors and do not necessarily rep- matography B, vol. 960, pp. 194–199, 2014. resent the official views of the NIH. [11] J. A. J. Haagensen, D. Verotta, L. Huang, A. Spormann, and K. Yang, “New in vitro model to study the effect of human Conflicts of Interest simulated antibiotic concentrations on bacterial biofilms,” Antimicrobial Agents and Chemotherapy, vol. 59, no. 7, All authors declare that they have no conflicts of interest. pp. 4074–4081, 2015. [12] P. A. Flume, P. J. Mogayzel Jr., K. A. Robinson et al., “Clinical Practice Guidelines for Pulmonary ,erapies C: cystic fibrosis Acknowledgments pulmonary guidelines: treatment of pulmonary exacerba- tions,” American Journal of Respiratory and Critical Care ,is work was supported by a grant (1R01 AI097380-0121) Medicine, vol. 180, no. 9, pp. 802–808, 2009. from the National Institute of Allergy and Infectious Dis- [13] D. J. Touw, A. J. Knox, and A. Smyth, “Population phar- eases of the National Institutes of Health and fund from the macokinetics of tobramycin administered thrice daily and 8 Journal of Analytical Methods in Chemistry

once daily in children and adults with cystic ﬁbrosis,” Journal of Cystic Fibrosis, vol. 6, no. 5, pp. 327–333, 2007. [14] C. Cheng, S. R. Liu, D. Q. Xiao, and S. Hansel, “,e application of trichloroacetic acid as an ion pairing reagent in LC-MS-MS method development for highly polar aminoglycoside compounds,” Chromatographia, vol. 72, no. 1-2, pp. 133–139, 2010. [15] J. Haagensen, D. Verotta, L. Huang, J. Engel, A. M. Spormann, and K. Yang, “Spatiotemporal pharmacodynamics of meropenem- and tobramycin-treated Pseudomonas aeruginosa bioﬁlms,” Journal of Antimicrobial Chemotherapy, vol. 72, no. 12, pp. 3357–3365, 2017. Hindawi Journal of Analytical Methods in Chemistry Volume 2018, Article ID 1308167, 24 pages https://doi.org/10.1155/2018/1308167

Review Article MS-Based Analytical Techniques: Advances in Spray-Based Methods and EI-LC-MS Applications

Federica Bianchi ,1 Nicolo` Riboni ,1,2 Veronica Termopoli ,3 Lucia Mendez ,4 Isabel Medina,4 Leopold Ilag ,2 Achille Cappiello,3 and Maria Careri1

1Department of Chemistry, Life Sciences, and Environmental Sustainability, University of Parma, Parco Area delle Scienze 17/A, 43124 Parma, Italy 2Department of Environmental Science and Analytical Chemistry, Stockholm University, 10691 Stockholm, Sweden 3Department of Pure and Applied Sciences, LC-MS Laboratory, Piazza Rinascimento 6, 61029 Urbino, Italy 4Instituto de Investigaciones Marinas, Spanish National Research Council (IIM-CSIC), Eduardo Cabello 6, 36208 Vigo, Spain

Correspondence should be addressed to Federica Bianchi; [email protected]

Received 22 December 2017; Accepted 26 February 2018; Published 23 April 2018

Academic Editor: Gauthier Eppe

Copyright © 2018 Federica Bianchi et al. (is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Mass spectrometry is the most powerful technique for the detection and identification of organic compounds. It can provide molecular weight information and a wealth of structural details that give a unique fingerprint for each analyte. Due to these characteristics, mass spectrometry-based analytical methods are showing an increasing interest in the scientific community, especially in food safety, environmental, and forensic investigation areas where the simultaneous detection of targeted and nontargeted compounds represents a key factor. In addition, safety risks can be identified at the early stage through online and real-time analytical methodologies. In this context, several efforts have been made to achieve analytical instrumentation able to perform real-time analysis in the native environment of samples and to generate highly informative spectra. (is review article provides a survey of some instrumental innovations and their applications with particular attention to spray-based MS methods and food analysis issues. (e survey will attempt to cover the state of the art from 2012 up to 2017.

1. Introduction analyze complex matrices. Interesting review articles and book chapters dealing with advances in ionization for mass Mass spectrometry (MS) is one of the most powerful spectrometry have been lately published [2–9]. techniques for the detection and identification of organic Recently, the advent of ambient MS technology paved and inorganic compounds. Being able to provide both the way for the development of a great variety of applications molecular weight and structural information [1], it is widely and innovations characterized by high throughput: the used in analytical laboratories for academic research, in- challenge of analyzing samples in their native state without dustrial product development, and regulatory compliance as sample treatment encouraged the development of new well as for proteomic or metabolomic studies, DNA char- techniques among which are the spray-based ionization ones acterization, drug discovery, environmental monitoring, including desorption electrospray ionization (DESI) [10], food analysis, forensics, and homeland security. paper spray ionization (PSI) [11], laser ablation electrospray A plethora of analytical MS-based methods based on the ionization (LAESI) [12], and easy ambient sonic-spray use of both stand-alone instruments and mass spectrometers ionization (EASI) [13]. coupled to different separation techniques such as gas and Novel materials and new instrumental configurations are liquid chromatography (GC and LC) or capillary electro- under study to enhance the performance of the different ion phoresis (CE) have been developed and validated in order to sources. Safety risks can be identified at the early stages 2 Journal of Analytical Methods in Chemistry through nontargeted monitoring technologies. Further- distinguished [23], and higher mass accuracy (even better more, the variety of fragmentation strategies that can be than 1 ppm). In addition, high mass resolving power allows combined in new instrumentation overall enhances work in discrimination between isobaric interference and ion of the omics fields, particularly proteomics and metabolomics. interest, leading to an accurate mass measurement even with Although MS-based methods are getting progressively a complex background. (ese features increase MS selec- more powerful, reliable, and easily available, the main tivity for the screening of nontargeted compounds in drawbacks are still related to sample complexity and complex matrices, providing a list of possible elemental preparation, mass accuracy, often requiring the use of high- compositions. Fu et al. have shown how important and resolution mass spectrometry (HRMS) to guarantee the efficient is the use of nontargeted screening with LC-HRMS univocal identification of the targeted compounds, and the to ensure quality and safety of food [24], whereas Mattarozzi need of high-throughput and screening analyses when and coworkers exploited the capability of HRMS for the a great number of samples have to be analyzed. rapid determination of melamine from melamine tableware (e aim of the proposed special issue is to cover the [25]. On the downside, HRMS-based methods generate aspects regarding emerging features of MS-based techniques complicated data that must be processed for “total ion focusing on innovative LC-MS studies and ambient MS with fragment spectra” to obtain high-quality mass spectral in- particular attention to the spray-based ionization tech- formation. Moreover, the mass of protonated or deproto- niques. New materials, prototypes, and instrumental con- nated molecules is not sufficient to prevent unambiguous figurations able to increase the performance of the developed compound identification. Hence, the use of spectra- methods will be presented and discussed. Finally, an over- matching approaches that utilize fragmentation ions could view of the most recent MS-based methods in food analysis be added to achieve additional information on the detected will be given. (is survey will attempt to cover the state of the compounds [26, 27]. False positives and false negatives are art from 2012 up to 2017. the major obstacles when screening complex samples. False negatives can occur due to very low concentrations, matrix 2. Advances in LC-MS interferences and suppression, and weak or no ionization. Due to these disadvantages, the practical application of LC- Electrospray ionization (ESI) is the technique of choice to MS and LC-HRMS is still far from the immediacy and produce ions suitable for mass analysis. ESI spectra typically simplicity of GC-MS. Taking into account that electron are characterized by single protonated or deprotonated ionization (EI) allows us to obtain characteristic and highly molecular ion (M + H)+, (M − H)−, and/or adduct ions. (e reproducible fragmentation of the analytes, a considerable low fragmentation is a limitation in compound character- effort has been devoted by the scientific community to in- ization through the use of reliable electronic libraries, crease compatibility between LC and EI-based MS to de- making necessary the use of multistage MS (MS/MS, MSn) or velop reliable, easy-to-use, and flawless interfaces. Moreover, HRMS to compensate the limited structural information. implementation of EI fragmentation to LC-amenable LC-ESI with triple quadrupole (QqQ) MS is the most used compounds could pave the way for many new fields of technique for qualitative and quantitative determination of research. targeted nonvolatile compounds in forensic and food ap- Recent developments in miniaturized mass spectrome- plications [14, 15]. In 2016, Remane et al. reviewed the ters have enabled these developments to be carried out to literature on applications of LC-MS/MS in clinical, forensic portable on-scene detection. In the next paragraphs, some of toxicology, and doping control since 2006 [16]. It must be the most popular and promising techniques are described: noted that ESI response is strictly affected by the mobile among them are LC-MS based on EI interfaces and spray- phase and sample composition as well as by the presence of based ionization techniques. coeluting interfering compounds, which may interfere with the ionization process. (ese phenomena are known as “matrix effects” (MEs) and can alter the response of the 2.1. LC-MS Based on Electron Ionization-Mass Spectrometry analytes causing either signal suppression or enhancement Interfaces. It is known that EI is ideal for the detection of [17, 18]. (e occurrence of ME introduces some critical a large number of GC-amenable compounds, but with an analytical shortcomings in quantitative analysis by LC-MS appropriate combination of several measures, it can become such as reduced sensitivity, nonlinear response, and low suitable for many LC-amenable small molecules having MW precision. In addition, the physicochemical properties of the up to approx. 600 Da. (ese compounds can be efficiently analytes can play an issue in the instrument response, thus converted into the gas phase, fast enough to be ionized introducing additional limitations. (e combination of before any decomposition process. powerful MS detectors with LC has solved many problems in For analysis of small-to-medium molecules, the coupling structural elucidation of unknown hazardous compounds between LC and EI-MS represents a valid strategy for [19]. In this context, HRMS is capable of providing full overcoming the main disadvantages related to ESI ionization spectral information by adding high mass resolving power and the use of costly and complicated techniques involving and accuracy to achieve selectivity and capability for ac- HRMS instrumentations. Furthermore, EI provides a rich curate mass measurements [20–22]. HRMS is characterized fragmentation pattern with a significant amount of struc- by higher mass resolution, defined as the mass difference tural information allowing a unique automated identifica- between two mass spectral peaks that can be clearly tion with structures at the isomer level [28]. Hence, the Journal of Analytical Methods in Chemistry 3 ability of EI for tentative identification of GC-amenable interface. It is designed to uncouple and separate the at- compounds is unparalleled even without HRMS. mospheric pressure found at the end of the HPLC system On the downside, the coupling between LC and MS with the high-vacuum zone of the ion source. (is specific based on EI represents a significant issue in the field of place is narrow enough to prevent vacuum from entering analytical chemistry. (e reason may be explained by the into the spray region allowing us to have an atmospheric antagonist conditions of operating. (e first one typically pressure zone where the vaporization process takes place. A works at ambient temperature and uses very high pressure removable silica-deactivated liner ensures a perfect con- for the efficient separation of the analytes, which are version into a gas phase before entering the mass spec- sometimes dissolved in a complex mobile phase. (e second trometer. A narrower fused silica capillary, called the inlet, one operates at a very high vacuum and high temperature. penetrates in the first portion of the liner and releases the LC (erefore, the effort of achieving and maintaining the high eluate. An inert gas flow surrounds the gas phase through the vacuum required for mass spectrometry is in contrast with vaporization microchannel and helps high boiling com- the intrinsic nature of HPLC, predominantly operating at pounds to vaporize. Figure 1 shows a complete layout of the high solvent flow rates. Also, the low tolerance of mass LC-MS system equipped with the LEI interface. spectrometers for nonvolatile mobile phase components (e rapid vaporization offered by the lined micro- contrasts with an HPLC dependence on nonvolatile buffers channel reduces the chance of thermal decomposition and to achieve high-resolution separations. capillary blockings, broadening the range of suitable ap- Since the year 2000, a few groups of researchers are plications, especially those regarding nontargeted analytes. working on the development of an efficient EI-based LC-MS Remarkable results were achieved in different conditions interface. and applications. Cappiello and his group played a significant role in the Over the past years, Seemann and his group developed innovation and improvement of LC-EI-MS interfacing and the supersonic molecular beams (SMB) LC-MS interface designing a series of systems characterized by steadily in- [34]. (eir studies started from the knowledge that standard creasing performance. Firstly, they presented a prototype of emission energy (70 eV) used in EI is not ideal for many the LC-MS interface called Direct-EI [29–31] based on direct NIST library compounds that have a weak (below 2% relative coupling of a low flow rate nano-HPLC with a high-vacuum abundance) or no molecular ion. (is issue is a critical point EI source. (e interface governs the direct introduction of when very large and thermally labile compounds are ana- a liquid-phase sample into the EI source of the mass lyzed. Furthermore, these analytes are usually less volatile spectrometer and the complete conversion of the liquid and require higher EI ion source temperatures with further effluent to the gas phase prior to a conventional electron- intra-ion source degradation, resulting in weaker molecular assisted ionization. (e core of the interface is represented ion production. To achieve a reliable EI-based sample by the nebulizer, which consists of a fused silica/PEEK identification, a more intense production of molecular ion is capillary, to guarantee a sufficient thermal insulation. (is needed. (us, the best ionization method should provide the interface was used in many different applications, not only in informative library searchable EI fragments combined with combination with chromatography but also in direct anal- enhanced molecular ions, especially for the compounds that ysis, as a universal detector for the targeted compound. are not included in the commercially available EI libraries. However, both nebulization and vaporization take place Taking into account these considerations, they present inside the ion source, leading to some drawbacks linked to a novel concept of the LC-SMB-MS system, based on the use capillary blockings. (is concern is mainly due to premature of supersonic molecular beams, as a medium for electron evaporation of the solvent making the analysis very difficult ionization of vibrationally cold sample molecules in a fly- under routine conditions. To meet the challenges of ana- through ion source. It is able to generate library searchable lyzing nontargeted compounds exploiting full potential of EI EI spectra and a more intense molecular ion. and the quantification of target compounds at low con- (e LC-SMB-MS apparatus is schematically shown in centration in complex matrices, Termopoli et al. presented Figure 2. a new, robust, efficient interfacing mechanism coming from A thorough evaluation of the interface, comprising the ground up [32]. (e new interface is called “liquid-EI” identification of unknown compounds using obtained li- (LEI). (e interface is completely independent from the rest brary searchable EI mass spectra, enhanced production of of the instrumentation and can be adapted to any gas molecular ions, demonstration of the absence of matrix chromatography-mass spectrometry system, as an add-on effects, simultaneous determination of semipolar and for a rapid LC-MS conversion. Secondly, with some little nonpolar compounds with reasonable detection limit, and tricks, it can be used with any HPLC system. Nanopumps low-cost instrumentation, was provided by that research and capillary pumps allow direct connection, and conven- group. (e group of Seemann demonstrates the feasibility of tional HPLC needs the use of a two-way splitter to reduce the the SMB interface as a valid tool in the analysis of unknown column flow rate to a level that is compatible with the in- compounds and as a low-cost LC-EI-MS system. terface, which is normally between 0.5 and 1 µl/min. In an A third group of researchers, headed by Rigano, pre- LEI interface, the vaporization of the LC eluate is carried out sented a new nano-LC-EI-MS for the determination of free at atmospheric pressure inside a suitable, independent fatty acids (FFAs) in mussels [34]. A selective and sensitive microchannel right before entering the ion source, called the nano-LC-EI-MS analytical method to investigate the FFA vaporization microchannel, representing the core of the profile in marine organisms and to monitor marine sentinels 4 Journal of Analytical Methods in Chemistry

Solvent Solvent B A

H e a He UV cell Agilent 1290 UHPLC pump t

Splitter Bypass Column Injector

Vaporization microchannel Metal

He Peltier Sample cooler He Removeable fused silica Fused silica capillary liner Figure 1: Global layout of the fully assembled system; the LEI interface, in gray, is between the UHPLC system and the MS detector. In the red circle, the vaporization zone is highlighted. Reprinted with permission from [33].

Q He gas 2 Nebulization Sheath agentCal. Make-up Triple quad LC Q1 Q3 MS SMB interface

Inj. DET. Q0 Column

Syringe pump Sample vaporization chamber Figure 2: EI-LC-MS with the SMB system outline. (e liquid is introduced either from the HPLC system after its column or from a syringe pump to the heated vaporization chamber through a pneumatic nebulizer. (e helium nebulization gas enters the SMB interface through a nebulization gas line, sheath gas line, and nozzle make-up gas line. Reprinted with permission from [35]. Journal of Analytical Methods in Chemistry 5 for the assessment of environmental pollution effects was pressure, without any tedious sample preparation steps or developed [35]. FFAs are minor components of the lip- laborious time-consuming chromatographic separation. idome, and they are usually analyzed by GC after a de- Technically, spray-based ionization techniques are based rivatization step, such as methylation or trimethylsilylation, on the use of electrospray droplets to extract the analytes is performed to convert FA into less-polar and more-volatile from the sample and transfer them to the mass spectrometer. moieties and improve their separation [9, 36]. However, the (e most common spray-based ionization technique is derivatization step, if not properly selected, can modify the DESI, in which a high-velocity pneumatically assisted ESI FA profile due to nonhomogeneous derivatization efficiency source generates charged microdroplets by the application of among different compounds (saturated, unsaturated, and a proper potential on the ESI needle. (e spray is directed polyunsaturated fatty acids). In addition, oxidation or towards the sample where the impact of the primary droplets isomerization products can be generated. Relative to this with the substrate leads to the formation of a micrometer- issue, LC can benefit over GC techniques from direct in- sized thin solvent film, able to solubilize the analytes at the jection of FFAs in their intact form, without any pre- liquid-solid interface. Secondary droplets containing the treatment. On the other hand, direct coupling with EI-MS analytes expelled by the film solvent are generated, and then, can benefit from the highly informative, repeatable, and desolvation and ionization in the gas phase occur, as in the reliable MS fragmentation. Drawing conclusions of the traditional ESI analysis. Finally, the ions are collected by the several attempts made by each group, they are following MS inlet. a distinctive pathway to obtain a common goal, the devel- In addition to DESI, several other techniques like nano- opment of a more useful and universal LC-EI-MS interface. DESI, EASI, and LAESI have been proposed. Probe elec- Regarding Direct-EI LC-MS, recent studies have been trospray ionization (PESI) is another interesting approach carried out also to increase the inertness of the electron based on the use of a solid conductive needle probe that ionization ion source by developing new materials [37, 38]. replaces the traditional electrospray capillary for sample As already stated, the vaporization surface of an electron introduction. Similarly, PSI is a technique based on the ionization MS source is a key parameter for the detection loading of the sample onto a triangular piece of paper from and characterization of targeted and untargeted analytes: it is which ions are generated by applying a high voltage in the known that difficulties in the vaporization process arise presence of a proper solvent [11]. Spray-based methods are when compounds characterized by high molecular weight suitable for the analysis of different compounds, from small and/or polarity have to be analyzed, thus requiring both the analytes, such as explosives [39–43], drugs [44–47], and food use of inert ion sources to reduce the interactions of the contaminants [48–50], to larger molecules such as lipids analytes with the stainless steel ion source and the use of high [51–53], peptides [54, 55], and proteins [56, 57]. source temperatures to promote analyte vaporization. In this field, Magrini et al. [37] proved that the use of a commercially available ceramic coating is able to improve the de- 3.1.Desorption ElectrosprayIonization-MassSpectrometry. DESI- tection of high molecular weight and high boiling MS is usually applied for surface desorption/ionization of compounds like polycyclic aromatic hydrocarbons (PAHs) analytes deposited on a probe material (PTFE, PMMA, glass, and hormones. More recently, Riboni et al. [38] were able to etc.) or directly from the sample surface. DESI-MS and DESI increase the inertness of the electron ionization ion source imaging have been successfully applied in different fields, by developing different sol-gel coatings based on silica, ti- such as forensic science [58], food control [59, 60], and tania, and zirconia. Again, the developed coatings were clinical applications [61–63]. tested for the Direct-EI LC-MS determination of PAHs and (e derivatization of metabolites deposited in solution steroids. (e best performances in terms of both signal peak onto a glass plate by dropping the derivatizing reagent on the intensity and peak width were obtained by using the silica- top of the dried analytes was proposed by Lubin et al. [64]. based coating, obtaining detection limits in the low ng/ml (e authors successfully applied this technique to several range with a good precision (RSD <9% for PAHs and <11% samples, demonstrating the possibility of performing mul- for hormones). No problems associated to ion cleaning were tiple and subsequent derivatization steps on the same spot. observed after prolonged use. An interesting approach was developed by Brown et al. [65] for the MS detection of fleeting reaction intermediates in electrochemical reactions utilizing a new waterwheel 3. Advances in Spray-Based working electrode setup. (e proposed technique allowed us Ionization Techniques to exploit DESI-MS operating at a low voltage. (e new apparatus consisted of a round rotating platinum working Nowadays, there is also a growing interest in the develop- electrode that was partially immersed in an aqueous elec- ment of real-time analytical technologies capable of allowing trolyte solution (Figure 3). During the rotation, a thin layer the direct detection of trace analytes in complex samples, of liquid film was deposited on the electrode surface, as in especially in their native environment. (e development of a waterwheel. A three-electrode system was set by using a new class of techniques, better known as “ambient ioni- a platinum wire counterelectrode and an Ag/AgCl reference zation techniques,” has introduced a revolution in the electrode, immersed in the reservoir of electrolyte solution. ionization field. (ese techniques are able to generate ions (e upper surface of the waterwheel was hit by a spray directly from native environment of the sample at ambient generated by a custom spray probe, thus allowing the 6 Journal of Analytical Methods in Chemistry

[67, 68]. In this approach, a PEEK capillary tube with Custom spray a micro-orifice is used to couple DESI with the UPLC source MS inlet capillary Stepper column. By using the proposed instrumental setup, a small motor amount of LC eluent (few nanoliters) is ionized by DESI with Ag/AgCl reference negligible time delay (6∼10 ms), whereas the remaining electrode analytes exiting the tube outlet can be collected. In addition, Pt counterelectrode online derivatization using reactive DESI is feasible increasing the charge of proteins and consequently enhancing Reservoir of the ionization yields. electrolyte An interesting novel configuration has been recently solution developed by Ren et al. [69]. (e authors developed Custom spray probe MS a method coupling slug-flow microextraction (SFME) and Pt working inlet Analyte capillary nanoelectrospray ionization for the MS analysis of organic Teflon disc electrode microdroplets compounds in blood and urine. A disposable glass capillary with a pulled tip for nano-ESI was used to perform the entire extraction and ionization process (Figure 4). Two adjacent liquid plugs were formed by injecting 5 µl of a proper organic Secondary solvent and 5 µl of body fluid (urine or blood) into the microdroplets capillary. Liquid-liquid extraction of the analytes was per- in film of formed by both moving the capillary and applying a push- electrolyte solution and-pull force through air pressure. After the extraction process, a high voltage is applied to the organic solvent plug Figure 3: Schematic representation of the developed experimental to generate the nano-ESI for MS analysis. setup. Reprinted with permission from [65]. (e proposed method was tested for the extraction and detection of different analytes, namely, methamphetamine, benzoylecgonine, verapamil, amitriptyline, epitestosterone, formation of secondary droplets, analyte ionization, and 6-dehydrocholesterone, 5α-androstane-3β, 17β-diol-16-one, their collection in the MS inlet. To avoid any electrochemical and stigmastadienone. Major analytical features were the oxidation or reduction on the electrode surface, no high reduced consumption of both the organic solvent and voltage was applied to the analyte spray, whereas a low sample. (e authors demonstrated that a direct de- potential (few volts) was applied to the platinum rotating rivatization of the extracted analytes in the organic phase electrode. was feasible, thus achieving excellent sensitivity with de- (e authors tested the new apparatus towards the detection limits in the 0.03–0.8 ng/ml range. tection of a diimine intermediates during electrochemical Despite its name, nanospray DESI (nano-DESI) is based oxidation of both uric acid and xanthine. on a different instrumental configuration compared to the An MS-electrochemistry coupling was also proposed by traditional DESI: its setup presents two different silica Looi et al. [66], who developed a new online electrochemistry- capillaries, one for solvent delivery and the other devoted to liquid sample desorption electrospray ionization-mass spec- the formation of charged liquid spray in front of the MS trometry (EC-LS DESI-MS) system. In EC-LS DESI-MS, an inlet. (e two capillaries are not in direct contact, thus electrosonic spray ionization source was used to generate producing a solvent bridge on the DESI surface. (e second a spray directed to the exit of the liquid sample capillary nanospray capillary produces a self-aspirating nanospray, positioned perpendicularly to the spray and the MS inlet. which is generated by applying a high voltage between the Separately, a thin two-electrode flow-through EC cell was MS inlet and the primary capillary. In comparison with connected to a syringe pump and was used to perform DESI, nano-DESI is characterized by higher efficiency in oxidation/reduction processes. (e ESSI-generated spray was liquid transportation and sampling performances. able to impact the outer surface of the LS capillary, which is (e capabilities of nano-DESI-MS were tested for the continuously coated by sample solution flowing at 200 μl/h, determination of pollutants and organic components in thus allowing the ionization of the analytes. (is prototype atmospheric fine particles by Cain et al. [70], in environ- was developed and tested using N,N-dimethyl-p-phenyl- mental aerosol by Tao et al. [71], and in clouds by Boone et al. enediamine (DMPA). Although oxidation of DMPA was [72]. In the clinical and pharmaceutical fields, both nano- already observed as a result of ionization of DESI-MS in DESI-MS and nano-DESI-MS imaging proved to be ex- positive mode, by applying a proper voltage to the online cellent techniques for the analysis of pharmaceuticals, electrochemical (EC) cell, it was possible to increase the yields biomolecules, and metabolites [73–77]. of the oxidation products, thus improving method sensitivity. A further instrumental innovation has been proposed Although DESI is usually coupled with high-resolution by Duncan and coauthors [78], who developed a pneumat- mass spectrometry, its coupling with LC is possible. A novel ically assisted nanospray desorption electrospray ioniza- splitting method for LC-MS applications, which allows both tion source. (e instrumental setup was based on the very fast MS detection of analytes eluted from the LC col- introduction of a secondary nebulizer replacing the self- umn and their online collection, was presented by Cai et al. aspirating secondary capillary in order to assist solvent Journal of Analytical Methods in Chemistry 7

HV MS Slug movement Bioﬂuid Extraction/spray sample solvent

Figure 4: Schematic representation of the SFME-nano-ESI sample processing. Reprinted with permission from [69].

ﬂow, to promote desolvation of the analyte, and to increase the distance between the nanospray and the MS inlet (Figure 5). (e developed device was tested for the analysis of rat kidney tissue sections, allowing us to obtain an improvement in sensitivity of about 1–3 orders of magnitude compared to the conventional setup. In addition, ion images characterized by high contrast, suitable for more intricate studies of metabolite distribution in biological samples, were obtained. A more complete desolvation of the analytes and reduced ionization suppression were additional features of the proposed device.

3.2.ExtractiveElectrosprayIonization-MassSpectrometryandLaser N2 Ablation Electrospray Ionization-Mass Spectrometry. Extractive N electrospray ionization (EESI) has been introduced in 2006 2 by Chen et al. [79]. It is based on the use of two different N2 sprayers: the ESI sprayer generates a charged solvent spray, whereas the sample sprayer has the function to nebulize the Solvent sample solution from an infusion pump. (e analytes are MS ionized in the collision area of the two sprays, and then, they are collected by the MS inlet. (e ionization mechanism of the EESI ion source was 0.16 mm ID 0.50 mm ID studied by Wang et al. [80], and different MS-based methods Teflon sleeve Teflon sleeve for the analysis of organic aerosols [81], drugs [82], pesti- Figure cides [83], amino acids [84], and biomarkers [85] in different 5: Picture and schematic representation of the pneumatically assisted nano-DESI ionization source. Reprinted with per- matrices were developed in the recent years. mission from [78]. LAESI-MS is another ambient ionization technique developed in 2007 by Nemes and Vertes [12]. Since most cells used for biomedical applications are cultured adher- Optical microscopy combined with LAESI-MS has been ently, the use of LAESI-MS was proposed to analyze ad- suggested by Compton et al. [87] in order to acquire both herent cells directly onto the culture surface, thus avoiding morphological and chemical information from tissue sam- chemical modification deriving from their detachment [86]. pling. In the developed instrumental setup, laser ablation In order to reduce the LAESI spot size, the authors applied occurred inside a chamber placed under an optical micro- a transmission geometry- (tg-) LAESI and incorporated an scope: the ablated particulates generated by the laser were objective with a high numerical aperture, thus achieving spot transported through a transfer tube by using nitrogen as sizes in the 10–20 μm range. (is technique (Figure 6) was carrier gas and finally ionized by the ESI spray. tested for the analysis of adherent versus suspended In order to compare the performances of the developed mammal cells, highlighting a difference in the metabolite prototype with those of the conventional LAESI-MS, plant compositions, thus proving that the cell detachment usually tissues were analyzed. In comparison with conventional LAESI, performed is able to produce chemical changes. On the the developed technique was characterized by reduced sensitivity contrary, tg-LAESI-MS allowed us to analyze directly the and dynamic range; however, these features were still sufficient cells in their native state and, due to the smaller spot size, to for the analysis and characterization of numerous metabolites reduce the sampled cell population by a factor of 20. and lipids in different spatial regions of biological tissues. 8 Journal of Analytical Methods in Chemistry

A novel paper spray cartridge with an integrated solid- phase extraction column has been developed by Zhang and Electrospray emitter w Manicke [92]. (e system was designed in order to perform on the same device the extraction, preconcentration, and θ ionization of the analytes from complex matrices such as blood or plasma. (e cartridge was divided into two parts: the bottom one containing the absorbent waste pad and the MS oriﬁce d paper spray substrate and the top one presenting a hole to host the solid-phase extraction (SPE) column. (e pro- Sample cedure for performing paper spray analysis is the following: (i) the samples are loaded onto the SPE column (sample Slide WD volume from 10 μl up to hundreds of microliters); (ii) the unbounded compounds pass through the SPE column and Reﬂective are absorbed by the waste paper pad; (iii) after sliding the top objective part of the cartridge to the paper spray substrate, the analytes retained on the SPE column are eluted and analyzed by PSI- Dichroic IR laser MS. (e analytical performances in terms of detection and mirror (λ = 2.94 μm) quantitation limits, recovery, and ionization suppression were evaluated for carbamazepine, atenolol, sulfamethazine, diazepam, and alprazolam. (e SPE cartridge allowed both Tube lens the selective enrichment of the targeted analytes from large sample volumes (up to hundreds of µl) and the removal of interfering compounds, thus enhancing signal intensities. Compared to direct PSI, the proposed method allowed us to Camera improve quantitation limits by a factor of 14–70, obtaining Figure 6: Schematic representation of tg-LAESI-MS. Reprinted limits in the 0.2–7 ng/ml range. with permission from [86].

3.4. Wooden-Tip Electrospray Ionization-Mass Spectrometry. Wooden-tip electrospray ionization-mass spectrometry (WT- 3.3. Paper Spray Ionization-Mass Spectrometry. In the last ESI-MS) is a rapid, in situ, and direct ambient technique based ten years, approaches based on the direct ionization from on the use of a wooden tip as a sampling and ionization solid substrates, such as paper spray, probe-based spray, leaf needle. (e tip is dipped into the liquid solution/matrix, and spray, and tissue spray, strongly increased. All these tech- after extraction, it is directly used as an ESI probe by applying niques are characterized by the generation of an electrospray a high voltage and spray solvent. (e analytes enriched on the directly from a probe. (e analytes in the samples can either tip are desorbed and ionized under ambient and open-air be ionized directly upon the substrate surface or be extracted conditions. on a probe and subsequently ionized within few minutes, (is method has been successfully applied by Yang et al. thus boosting analysis speed. [93] for the analysis of pesticides, toxicants, date rape drugs, Innovations in PSI-MS have been described by Duarte and illicit additives in various food samples. (e capabilities et al. [88] and Salentijin et al. [89] who developed 3D-printed of untargeted WT-ESI-MS analysis for the identification of cartridges in order to obtain a solvent reservoir, thus the sources of plant materials by using a multivariate statistic allowing us to prolong the spray generation from the paper approach were exploited by Xin et al. [94], whereas Yang and tip. A supporting prototype able to automatically perform Deng [95] used an internal standard WT-ESI-MS-based PSI-MS analysis of a great number of samples, suitable for method to obtain the fingerprint mass spectra for rapid high-throughput applications, has been designed by Shen quality assessment and control of Shuang-Huang-Lian oral et al. [90]. Finally, the coupling of SFME with PSI-MS for the fluid, an herbal preparation registered by Chinese Phar- rapid analysis of macrolide antibiotics at the trace level in macopoeia. By using the internal standard and principal biological samples such as whole blood, milk, and other body component analysis (PCA), it was possible to obtain the fluids has been proposed in a study carried out by Deng et al. fingerprints of samples from different manufacturers. A [91]. (e same approach was applied for the detection of bamboo pen nib shaped and used for sample loading and an perfluorooctanesulfonic acid and perfluorooctanoic acid ESI probe for the determination of 4-chloro-amphetamine from Daphnia magna body fluids. After the SFME extrac- was developed by Chen et al. [96], resulting in lower de- tion, the organic extract was simply spotted on the PSI paper. tection limits compared to PSI-MS and traditional WT-ESI- Excellent results were achieved in terms of linearity range MS analyses. (5–500 and 0.5–50 ng/ml range for antibiotics and per- Similarly, a WT-ESI-MS method combined with dif- fluorocompounds, resp.) and sensitivity (LOQs 0.3–1.3 and ferent nonpolar solvents for the detection of native proteins 0.03–0.30 ng/ml for antibiotics and perfluorocompounds, and protein complexes directly from raw biological samples resp.). Recovery rates always higher than 85% were obtained. has been proposed by Hu and Yao [97]. Journal of Analytical Methods in Chemistry 9

(e applicability of field-induced wooden-tip electro- Further experiments were carried out by deposing the spray ionization-mass spectrometry (FI-WT-ESI-MS) for sample on the probe surface while maintaining the EDTA in high-throughput analysis of herbal medicines has been the ESI spray solvent (functional electrospray). Excellent proved by Yang et al. [98]. Field-induced ESI was performed results in terms of sensitivity were achieved, thus proving by applying a high voltage on the MS inlet, thus allowing the method reliability for the rapid analysis of metal substrates creation of a strong electric field between the capillary without sample preparation. emitter and MS inlet to induce ESI from the sample solution. An electrostatic spray ionization (ESTASI) method for A high-throughput analysis device was developed by the the analysis of samples deposited in or onto an insulating application of sample-contactless high voltage on the MS substrate has been proposed by Qiao and coworkers [102]. In inlet. In addition, the switch between positive and negative this study, the ionization of the analytes is induced by ion detection modes can be readily accomplished, thus a capacitive contactless coupling between the electrode and providing complete MS information of the analyzed sam- the sample: by applying a pulse high voltage to the electrode, ples. (is approach allowed us to boost the analysis speed: 6 s an electrostatic charging of the sample occurs, leading to per sample was sufficient to perform the analyses. a bipolar spray pulse. When the applied voltage is positive, (e proposed method was applied for the rapid de- the bipolar spray pulse consists first of cations, followed by termination of various active ingredients in different raw anion production. (e instrumental setup can be modified herbs and herbal medicines. (e obtained mass spectra were in order to obtain ion emission from samples in a silica used as fingerprints for tracing the origins, establishing the capillary, in a disposable pipette tip, and in a polymer authenticity and assessing the quality of herbal medicines. microchannel or deposited as droplets onto an insulating Very recently, a novel and noninvasive sampling method poly(methyl methacrylate) plate presenting wells or hy- using a watercolor pen (brush) rinsed with ethanol as drophilic patches. (is technique proved to be suitable for a sampling tool to collect analytes from the eyelids of vol- the analysis of peptides and proteins. unteers has been evaluated [99]. (e brush was placed between the mass inlet and the ESI plume, thus allowing the 4. Materials for Spray-Based desorption and ionization of the analytes. (e results achieved proved the suitability of the developed technique Ionization Techniques for the semiquantitative determination of caffeine and its (e development of new materials is a field of increasing metabolites in eyelid samples. interest in order to enhance the performances of analytical methods. Both high selectivity and increased ionization efficiency are demanded to detect analytes at trace levels in 3.5. Miscellaneous. Following the method developed by Pan complex matrices and to shorten analysis time. Different et al. [100], based on the use of a single probe inserted into studies have been published dealing with the development of a single cell for sampling intracellular compounds by real- novel surfaces for ambient MS. A brief overview on the most time MS analysis, Chen et al. [52] described a novel method recent materials developed for spray-based ionization for single cell analysis and lipid profiling by combining drop- techniques with particular attention to DESI-MS and PSI- on-demand inkjet cell printing and probe electrospray MS applications is described in the next paragraphs. ionization-mass spectrometry (PESI-MS). Droplets containing single cells were generated from a cell suspension by inkjet sampling, precisely dripped onto the tungsten tip of 4.1. Materials for DESI-MS. In 2005, Takats et al. [103] the ESI needle, and sprayed under a high-voltage electric demonstrated that the DESI ion source is strongly influenced field. Cellular lipid fingerprints were then obtained by MS by the dielectric constants of both the substrate material and detection. (e analytes from eight types of cells were de- the spray. tected, and PCA analysis was performed in order to dif- (e effect of surface chemistry in the DESI ionization ferentiate the samples. (e proposed platform was mechanism has been deeply investigated by Penna and demonstrated to be suitable for the direct MS profiling of coworkers [104]: the performances of different glass sub- single-cell lipid species without derivatization or the labeling strates obtained by the sol-gel technology and functionalized procedure. by using different alkylsilanes were tested and compared in A method for the direct characterization of metals in terms of ionization efficiency. (e substrates were charac- solid samples using electrospray laser desorption ionization- terized in terms of surface free energy and wettability. Owing mass spectrometry (ELDI-MS) has been developed by Shiea to their different polarity, melamine, tetracycline, and lin- et al. [101]. (e main advantages over classic approaches comycin were used as model compounds. A significant were related both to the absence of sample pretreatment and decrease in the ionization efficiency was observed when to the presence of very short analysis time. Mixtures of more hydrophilic surfaces were used, thus demonstrating metal-EDTA complexes were applied on a stainless steel the pivotal role of both hydrophobicity and wettability to plate and submitted to ELDI-MS analysis. (e capabilities of increase the performances of DESI-MS experiments. the technique were initially tested by spotting the metal- More recently, a 3D printed polylactic (PLA) supports in EDTA complexes on a solid probe and performing laser order to detect insulin and gentamicin in chitosan gels have ablation of the material. (e ablated analytes, present as been proposed by Elviri and coauthors [105]. By using 3D EDTA complexes, were ionized in the electrospray plume. printing, hemispherical wells were created, thus allowing us 10 Journal of Analytical Methods in Chemistry

Solvent Acetonitrile/H2O MS inlet spray DESI

Spotting On-surface digestion Model proteins Protein solution LESA ESI chip Trypsin solution

Calibration solution Scan direction 5 mm

Spotting Extraction Meat In-solution digestion solvent

Figure 7: Schematic representation of the procedure proposed by Montowska et al. Reprinted with permission from [106]. to obtain DESI-MS responses five times higher than those functionalization of the inner wall of the sample transfer achieved by using PTFE commercial slides. Improvement in capillary with enzymes, thus allowing the fast and online terms of signal stability was also achieved, thus suggesting digestion of proteins. the capability of 3D printing technology to improve the Noticeable applications of DESI-MS are based on its desorption step in DESI-MS. coupling with new sampling devices and extraction and Novel substrates have been proposed also for proteomics separation techniques to develop methods for high- and peptidomics: the capability of Permanox slides for both throughput analyses. DESI-MS and liquid extraction surface analysis-mass (e microfabrication of ultrathin-layer chromatography spectrometry (LESA-MS) of skeletal muscle proteins ob- (UTLC) plates via conformal low-pressure chemical vapor tained from a mixture of standard proteins and raw meat has deposition of silicon nitride onto patterned carbon nanotube been discussed in a recent study [106]. (e proposed method (CNT) scaffolds, acting as surface templates, has been is schematically reported in Figure 7. described by Kanyal et al. [109]. (e plates were heated and In both cases, good responses were obtained with LESA-MS oxidized to both remove the CNTs and convert Si3N4 into characterized by higher sensitivity and stability with respect to silica; finally, the plates were hydroxylated in aqueous am- the DESI-MS approach. Finally, multivariate data analysis monium hydroxide. (e resulting UTLC phases did not show allowed the correct discrimination among different meat classes. any distortion of the microfeatures and were characterized by Rapid and simple analyses represent a key parameter in a higher robustness in comparison to high-pressure TLC proteomics: an interesting study was carried out by Dulay et al. plates. Good results in terms of chromatographic perfor- in 2015 [107]. Two hybrid organic-inorganic organosiloxane mances were observed obtaining a faster separation when polymers functionalized by immobilized trypsin (T-OSX) for mixtures of lipophilic, water-soluble, and fluorescent dyes are the on-surface and in situ digestion of four model proteins, to be analyzed. A strong reduction in terms of mobile-phase that is, melittin, cytochrome c, myoglobin, and bovine serum consumption and an enhanced lifetime were observed. Fi- albumin, were developed and tested under DESI-MS and nally, the UPTLC plates were submitted to both DESI-MSI nano-DESI-MS conditions. (e silica polymers were obtained and direct analysis in real-time (DART)-MS analyses, via sol-gel technique using methyltrimethoxysilane and showing a good compatibility with common ambient de- dimethyl-dimethoxysilane as precursors. (e OSX polymers sorption and ionization techniques. were derivatized with trimethoxysilylbutyraldehyde and In the same year, Ewing et al. [110] described the DESI- functionalized with different amounts of trypsin. In both cases, MS detection of the low vapor pressure nerve agent simulant despite the low enzyme-to-substrate ratios, the achieved re- triethyl phosphate. (e analyte was previously adsorbed sults proved the suitability of the developed substrates to allow onto silica gel, forming a very fine particulate that was on-surface protein digestion followed by direct DESI-MS collected by using a sticky screen sampler. (e device was analyses obtaining sequence coverages in the 65–100% range. characterized by a stainless steel screen presenting a partially Taking into account that DESI-MS analyses can be polymerized polydimethylsiloxane (PDMS) coating. (e performed also on liquid samples, an improvement of the quantitative collection of the particulate sample from apparatus commonly used for this kind of analyses has been a contaminated surface was achieved by interfacing the proposed [108] by replacing the sample transfer silica sticky screen sampler with a bioaerosol collector. Finally, the capillary with a trap column filled with chromatographic sticky screen was placed onto a moveable platform mounted stationary phase materials, such as C4 and C18. (e proposed in front of the DESI-MS instrument, thus allowing a re- system proved to be suitable for trace analyses of both organic producible sample introduction system able to minimize compounds and biomolecules such as proteins/peptides in sampling errors. DESI-MS analyses performed directly on complex matrices in the presence of high salt content. the PDMS coating allowed us to obtain a very low detection Another interesting feature was related to the covalent limit suitable for trace detection. Journal of Analytical Methods in Chemistry 11

Electrospun nylon-6 nanofiber mats for DESI-MS a region for inducing the sample inlet into the mass spec- analysis and imprint imaging have been developed by trometer. (e proposed stamping method allowed rapid Hemalatha et al. [111]. (e nanofibers were developed by prototyping of microfluidic paper-based analytical devices, needleless electrospinning: nylon-6 was dissolved in formic without the need of sophisticated instrumentation. Different acid and the solution was electrospun at room temperature. types of papers were investigated: an increase of the PSI-MS Uniform mats of varying thicknesses composed of ∼200 nm responses of xylose and glucose as a function of the decrease diameter fibers were obtained: the properties of these ma- of porosity of the paper substrate was observed. PS-PB terials can be tuned by varying spinning conditions and showed the best performance compared to the conven- surface functionalization. As model compounds, dyes and tional paper and paper with two rounded corners. PS-PB was the extract of periwinkle flower were spotted onto the nylon applied to detect sugars and their inhibitors in liquors from nanofiber mat, thus obtaining a uniform coating of the fi- a second-generation ethanol process, thus obtaining excel- bers. DESI-MS analysis allowed us to obtain spectra without lent results in terms of linearity (over two orders of mag- polymer interference and reproducible DESI-MS images. nitude) and limits of detection and quantification. (e authors also demonstrated that compounds of interest Another interesting study carried out by Zhang et al. could be incorporated into nanofibers during their forma- [118] proposed the use of commercially available silica- tion by using as model compounds the crude methanol coated paper in order to increase the PSI-MS responses of extract of periwinkle flower and tetraphenylphosphonium therapeutic drugs in dried blood spots. (e presence of silica bromide (TPPB). (e major metabolites of periwinkle were gel particles in the cellulosic framework of the silica-coated identified by DESI-MS, even though the spectrum was paper produced a reduced diffusion of the blood through the different compared to that obtained by spotting the sample. substrate, thus leading to a higher percentage of blood TPPB was detected with no nylon interference. (e authors sample available on the top side of the substrate. By oper- demonstrated the possibility of imprinting patterns made of ating under the optimized conditions, that is, by using printing inks, plant parts, and fungal growth on fruits on the dichloromethane/isopropanol (9 :1 v/v) as a spray solvent nanofiber mats. Metabolites were identified by DESI-MS. mixture, limits of quantitation of about 0.1 ng/mL were (e results highlighted that electrospun nanofiber mats achieved, with a sensitivity gain of 5–50-fold in comparison could be considered as smart surfaces to capture diverse to chromatography papers. classes of compounds for rapid detection or to imprint CNT-coated filter paper for low-potential PSI-MS imaging under ambient conditions. analysis of different organic molecules has been used by Narayanan and coauthors [119]: by applying a voltage of 3 V, full-range mass spectra similar to those obtained by con- 4.2. Materials for PSI-MS. Although traditional paper spray ventional ESI at 3 kV could be recorded. (e advantage of ionization is performed on the filtering and chromato- the proposed material relies on the use of very mild con- graphic paper [11, 112–115], researchers have focused their ditions for the ionization of the analytes, thus allowing the attention to the development of new functionalized sub- detection of compounds that could be easily oxidized. (e strates in order to obtain substrates characterized by en- performances of the proposed analytical method were hanced selectivity and sensitivity. assessed for a wide range of volatile and nonvolatile com- Very recently, Lai et al. [116] compared the ionization pounds, such as amino acids, antibiotics, and pesticides in performances of different paper-like substrates obtained different matrices. from both natural fibers and synthetic fibers, namely, gampi Very recently, Wei et al. [120] synthesized graphene oxide paper, Tengujou paper (natural), polycarbonate, polylactic (GO) nanosheet-modified N+-nylon membrane (GOM) for acid, and poly-l-lactic acid (PLLA) (synthetic), with those of the extraction of malachite green (MG), a highly toxic dis- traditional chromatographic paper for the analysis of de- infectant, and its metabolites in liquid samples and fish meat. signer drugs. (e surface characterization of the developed GO nanosheets are characterized by a very high surface area materials showed the presence of different surface mor- (∼800 m2/g), suitable for MG adsorption via π-π stacking and phologies able to affect PSI capabilities: gampi paper and electrostatic interactions. GOM was obtained by self- PLLA nanofibers, characterized by a tough and extremely deposition of GO thin films onto N+-nylon membranes. thin structure, were able to promote signal enhancement, (e material was tested both as a direct spray ionization thus allowing us to reach lower limits of detection. (ese substrate and as an LDI-MS probe. (e latter application findings could be explained by taking into account the re- resulted in no significant response, whereas the coupling duced thickness of the used papers: by operating under the between GOM and direct spray ionization allowed the described conditions, a rapid evaporation of the sample quantitation of MG and its metabolites at nanomolar levels molecules occurs, thus increasing the speed of the ionization with extraction recoveries higher than 98%. process. An improvement of the performances of PSI-MS in (e analytical performances of paper with paraffin terms of sensitivity has also been achieved in a study dealing barrier (PS-PB) for the PSI-MS detection of hydrox- with the use of a paper substrate functionalized with urea ymethylfurfural (HMF) and sugars like glucose and xylose in [121]. Triangles of chromatography paper were treated with sugarcane liquors have been tested by Colletes et al. [117]. 1-[3-(trimethoxysilyl)propyl]urea to create an anion capture Microfluidic hydrophobic channels were obtained using layer. (e authors demonstrated that the urea-modified paraffin barriers on paper substrates, thus delimiting paper is an excellent substrate for PSI-MS since it is able 12 Journal of Analytical Methods in Chemistry to reduce ionization suppression caused by anions and explained, thus allowing us to predict detection limits very highly polar compounds in the negative-ion MS mode. closed to those observed experimentally and to calculate the (ese findings are of pivotal importance for the analysis of relative surface activity of both positive and negative ions. biological samples like urine, blood, and plant extracts. A selective substrate for PSI-MS, based on the use of 4.3. Materials for Wooden-Tip Electrospray Ionization-Mass molecularly imprinted polymers (MIPs), has been proposed Spectrometry. Surface-coated wooden-tip electrospray by Pereira et al. [122]. More precisely, a membrane spray ionization-mass spectrometry (SCWT-ESI-MS) is a new ionization method, combining MIP extraction and PSI-MS technique based on the use of a functionalized wooden analysis, was developed and tested for the determination of needle, acting both as extraction/enrichment phase and ESI diuron and 2,4-dichlorophenoxyacetic acid from apple, probe. By using this strategy, the tip is coated by a proper banana, and grape methanolic extracts. Being used as PSI sorbent for highly selective enrichment of targeted com- substrates, MIPs were synthesized directly on a cellulose pounds from complex matrices, thus making it suitable for membrane: the bulk of the MIP was made by ethylene analyses at ultratrace levels. Luan suggested the use of glycol dimethacrylate, using monuron and 2,4,5-tri- a SCWT-ESI-MS technique to detect different analytes in chlorophenoxyacetic acid as templating agents. After ex- complex matrices [125, 126]. traction, the MIP membranes were washed to remove matrix (e SCWT-ESI-MS technique has been applied for the interferences and tested as PSI-MS substrates using meth- detection of perfluorinated compounds (PFCs) in complex anol as a spray solvent. (e use of these novel materials environmental and biological samples at ultratrace level [125]. allowed us to obtain signal intensities of the targeted analytes Sharp wooden tips were functionalized via the silanization far higher than those obtained by nonimprinted polymers process by using octadecyltrimethoxysilane and n-octade- with detection limits in the µg/l range. cyldimethyl[3-(trimethoxysilyl)propyl]ammonium chloride Bills and Manicke [123] developed a disposable paper (OTPAC), in order to obtain two different extractive phases, spray cartridge containing a plasma fractionation membrane characterized by long alkyl chain. (e two phases were tested to perform on-cartridge plasma fractionation from whole for the extraction of PFCs spiked water. (e OTPAC coating blood samples. (ree commercially available blood frac- was characterized by the best enrichment capabilities: the tionation membranes, made of different materials, ranging extraction is performed not only by the reversed phase, but from polymers to natural and synthetic fibers, that is, Vivid also by exploiting the ion exchanged adsorption mechanism. polysulfone membrane, NoviPlex plasma fractionation card, Morphological studies of the tip showed a high probe po- and CytoSep, were tested. Even though all the materials were rosity, thus increasing the surface area of the material, and capable of interacting with plasma samples with low levels of presence of microchannels for transport of the solvent to cell lysis, difficulties in terms of drug extraction were ob- achieve ambient ionization MS analysis. After method opti- served. In particular, Vivid polysulfone membrane and mization, the probe was tested for the detection of eight NoviPlex plasma fractionation card exhibited a high binding different PCFs both in pure water and in complex matrices, capability (over 30%) for all the tested drugs, whereas that is, lake water, river water, whole blood, and milk. (e CytoSep showed a lower binding affinity (<17%) only for achieved results proved that the SCWT probe is characterized two out of five drugs. A drawback of the developed device by outstanding enrichment capabilities, thus being able to was also related to the poor fractionation efficiency, as enhance method sensitivity by approximately 4000–8000- measured by the red blood cell content in the fractionated folds and 100–500-folds in aqueous samples and in whole plasma. Quantitative analysis of plasma using PSI-MS blood and milk samples, respectively. Method validation provided results closed to those obtained by HPLC-MS resulted in good linearity (two orders of magnitude), excellent without the need of offline extraction or chromatography quantitation limits (in the 0.21–1.98 ng/l range), and accuracy separation. (recovery rates in the 89–112% range). A new zero volt-paper spray ionization (ZV-PSI) has been Similarly, a SCWT-ESI-MS-based method has been developed recently by Wleklinski et al. [124]: this approach is tested for the rapid and sensitive analysis of trace fluo- based on the generation of the electrospray by the action of roquinolone and macrolide antibiotics in water [126]. (e the pneumatic force of the vacuum at the MS inlet. ZV-PSI wooden probe was functionalized via silanization and sul- analyses were performed over a large variety of samples, fonation reactions in order to obtain a sulfo-C8-chain including tributylamine, cocaine, terabutylammonium io- coating able to interact with the analytes with both reversed dide, 3,5-dinitrobenzoic acid, fludioxonil, and sodium tet- phase and ion exchange mechanisms. (e SCWT-ESI-MS raphenylborate. In comparison to classic PSI-MS, the method was then optimized and tested for the extraction of achieved results showed a strong decrease of signal intensities four fluoroquinolone and two macrolide antibiotics in water for all the investigated analytes. Although the range of ana- at trace levels. Method sensitivity allowed us to obtain de- lytes useful for ZV-PSI-MS analysis resulted to be very similar tection and quantitation limits in the 1.8–4.5 and 5.9– to standard PSI-MS, differences in mass spectra were ob- 15.1 ng/l range, respectively. Again, linearity was verified tained. (e observed behavior was related to the ionization over two orders of magnitude: good precision (RSD <14.3%) mechanism of the proposed approach, which is strongly and recovery rates in the 93.6–112.6% range were other related to the effects of analyte surface activity. By using features of the developed method. Finally, the developed a Monte Carlo simulation, the mechanism regarding the method was applied for the analysis of the targeted anti- formation of ions from initially uncharged droplets was also biotics in tap and river water samples. Journal of Analytical Methods in Chemistry 13

SMAMIE Template 120°C removal Wooden tip polyreaction MIPCWT-SPME probe

Spray solvent

LMG Extraction Desorption

HV MG

MS inlet Figure 8: Schematic representation of the instrumental setup proposed by Huang. Reprinted with permission from [127].

An interesting approach based on the use of molecularly characterized by very different polarity. After sample sep- imprinted polymers as coating for SCWT-ESI-MS has been aration, the hydrophobic substrate was coated with chlor- discussed in a recent study [127]. (e coating was synthe- otrimethylsilane to form hydrophobic surfaces, suitable for sized by applying a silicone-modified acrylate molecularly ESTASI analysis. Both TLC plates were considered ideal imprinted emulsion onto the tip surface (Figure 8). (e substrates for in situ characterization of samples by using developed material was tested for the detection at trace levels ESTASI-MS, with efficient analyte extraction and separation. of malachite green and its metabolite leucomalachite green In addition, the capability of removing interfering com- in aqueous samples. (e MIP-SCWT probe exhibited high pounds such as salts increased method performance, thus enrichment capabilities, allowing us to obtain detection allowing the detection of the investigated analytes at trace limits at low µg/l levels. In addition, a good linearity was levels. obtained for both the compounds (three orders of magnitude). (e method proved to be suitable for high-throughput analysis and was successfully applied for the analysis of tap 5. MS-Based Approaches for Food Analysis water, river water, and fish samples. (e demand for safe and high-quality foods has significantly increased in recent years. Food safety and quality have 4.4. Miscellaneous. Similar to PSI, aluminum-foil mass become of greater importance, and the governments of spectrometry (Al-ESI-MS) was recently developed by Hu many countries have increased the amount of relevant et al. [128]. (is technique is based on the use of a household legislation and demands for food authentication [130]. In aluminum foil to obtain the spray ionization of the analytes. consequence, the development of more robust, efficient, (e Al foil was cut into triangles, which were folded sym- cost-effective, and powerful analytical methodologies is metrically to obtain a mini-reservoir for the sample solution, continuously needed in order to face these requirements. MS and was connected to the high voltage supply of the mass is one of the most suitable techniques because it is featured spectrometer. (e proposed technique was tested for the by excellent specificity, sensitivity, and throughput [131]. MS direct analysis of a wide variety of complex matrices, namely, has been widely used in food safety and quality analysis, and energetic beverages, urine, skincare and medical creams, and recent advances in MS can provide faster and more accurate herbal medicines. (e inert, hydrophobic and impermeable methods able to offer better qualitative and quantitative surface of the Al foil allowed effective on-target extraction of results. Additionally, coupling mass analyzers with separa- solid samples and on-target sample clean-up, that is, re- tion techniques, such as liquid chromatography (LC-MS) moval of salts, adulterants, and detergents from proteins and and gas chromatography (GC-MS), have significantly im- peptides. Being Al an excellent heat conductor, the direct proved food analysis for screening, identification, structural monitoring of thermal reactions, such as thermal de- characterization, and quantitation purposes. naturation of proteins, can be performed in an easy way by One of the most challenges in the application of MS in Al-ESI-MS. food analysis, especially in detection of contaminants, is In a different research study, ESTASI was applied to sample preparation because foods are considered very identify and quantify different compounds from silica gel complex matrices in which some natural components can surfaces, via direct coupling with TLC [129]. (e sample negatively influence the analysis of the targeted compounds. spots separated by TLC were analyzed by ESTASI-MSI. (e Traditional methods for sample extraction include solid- analyses were performed on both drug molecules, using liquid extraction (SLE), liquid-liquid extraction (LLE), and normal phase TLC, and dyes using reversed phase C18 TLC solid-phase extraction (SPE). More recent is the use of solid- plates, thus guaranteeing the analyses of compounds phase microextraction (SPME), pressurized liquid extraction 14 Journal of Analytical Methods in Chemistry

(PLE), and QuEChERS (quick easy cheap effective rugged Almost all these applications combine QqQ-MS with LC safe) [132]. (e introduction of high-resolution mass or GC separation methods. In some cases, LC- and GC- spectrometers, which provide extremely high selectivity and based techniques were also coupled with other types of MS sensitivity, or other emerging MS strategies such as ambient- analyzers such as ion traps (ITs) or quadrupole-linear ion ionization MS, direct food analysis, and matrix-assisted traps (Q-LITs), TOF, or Q-TOF to determine food con- lased desorption ionization-time of flight-mass spectrome- taminants [166–168]. try (MALDI-TOF-MS) profiling and imaging, has strongly Multidimensional procedures allowed us to increase re- reduced sample preprocessing. solving power and separation capabilities that can be bene- According to the PubMed database, since 2012 about 20000 ficial for subsequent MS-based detection, considering that the publications dealing with developed MS-based methods for targeted compounds can reach the detector more separated in food safety and quality purposes are available. In this section, time. (is is the case of comprehensive two-dimensional gas we do not intend to provide an exhaustive revision of all chromatography (GC × GC) that has been coupled to a TOF- published studies, but an overview of the most important MS MS analyzer to determine dioxin-related pollutants in techniques proposed to evaluate food safety and quality. complex food samples [169], to screen 68 pesticide residues in oilseed [170], or to detect and to quantify different poly- chlorobiphenyls (PCBs), polybrominated diphenyl ethers 5.1. MS-Based Approaches for Food-Safety Assessment. (e (PBDEs), and PAHs in fish samples [171]. main purpose of food analysis is to ensure food safety, thus More recently, HRMS analyzers, typically Q-TOF, requiring the development of accurate and reliable meth- Orbitrap-MS, and Fourier-transform ion cyclotron reso- odologies for the detection of microbial-related spoilage, nance- (FT-ICR-) MS, have been used in the field of food determination of allergens, detection of environmental safety. (ese instruments are characterized by high reso- contaminants as well as banned external compounds, or the lution (100,000–1,000,000 FWHM) and high mass precision assessment of the occurrence of natural toxins. (ese methods (1-2 ppm, allowing discrimination between isobaric in- are strongly influenced by current legislation, which estab- terferences and ions of interest), thus making possible the lishes the requirements that an analytical method must meet screening of unknown compounds with a full MS scan and for an unequivocal identification and quantification of the construction of databases for targeted compounds. For a controlled substance in food samples, as well as the max- instance, UPLC-Orbitrap-MS was used to create a database imum residue limits (MRLs) on certain substances [133]. of more than 350 compounds in honey [172]. (ese data- Being able to allow the quantification of known com- bases included different classes of pesticides and veterinary pounds with great selectivity and sensitivity, tandem MS drugs and allowed simultaneous screening of analytes and detection is one of the most frequently utilized analytical identification and quantitation of detected compounds in approaches to determine contaminants in foods. Triple different honey samples. Similar UPLC-HRMS approaches quadrupole (QqQ) mass spectrometers, running under have been lately used to create an accurate-mass database multiple reaction monitoring (MRM) mode, are the most including the fragmentation of more than 600 different food popular instruments for detecting contaminants in food. contaminants, such as pesticides, veterinary drug residues, (is detection procedure allows us to verify the compliance mycotoxins, and perfluoroalkyl substances [173]. Since the with European legislation on banned and controlled sub- particle size of the stationary phase in UPLC is significantly stances in foods [133]. lower than that observed in HPLC, UPLC yields higher Since 2010, numerous researches have used this meth- speed, better resolution, increased sensitivity, and better odology to detect pesticides in several fruits and vegetables, peak capacity. Additional examples are related to the de- such as tomato [134–137], orange [138], mandarin [139], rice velopment of LC-Orbitrap-MS-based methods for pesticide and red pepper [139–141], avocado [142, 143], apples and screening in vegetables and fruits [174], as well as for the cucumbers [144, 145], mango [146], tea [147], lettuce [148], analysis of 18 selected mycotoxins in baby food [175]. grains and cereals [149–152], soybean products [153], Ambient MS-based techniques have also been widely groundnut oil [154], and wines [155]. (e same method- applied for food safety purposes: different ionization tech- ology was also used in the identification and quantification niques have been used like LAESI-MS to detect neurotoxin of veterinary drug residues in shellfish [156], meat [157, 158], domoic acid in shellfish [176], DESI-MS for the rapid detection eggs and milk [158], and contaminants from food contact of shellfish poisoning toxins in mussels [39], and PSI-MS for materials [159]. the determination of pesticides in fruits and vegetables [114]. Triple quadrupoles in MRM mode is also the most- extended approach to detect and quantify toxins and pathogens in food. (ese pathogens can contaminate foods 5.2. MS-Based Approaches for Food Quality Assessment. Besides directly or indirectly, through the productions of toxins. (e food safety, food quality is one of the main concerns of the control of toxin and pathogen levels is extremely important, modern food industry. Food quality encompasses multiple since the consequences on health due to their contamination factors, since food authentication and adulteration of food of food may be very serious. Following this approach, food characteristics include food ingredients, such as lipids, products such as nuts [160], maize [161], shellfish [162], proteins, oligosaccharides, vitamins, and carbohydrates, and tomato [163], beer [164], and multicereal baby food [165] additives, such as preservatives, antioxidants, and chemicals were analyzed. used for flavor, color, and odor. As a consequence, the Journal of Analytical Methods in Chemistry 15 evaluation of food quality usually represents a very complex perception of rice aroma [202]. (ese results are valuable for task that needs to consider multiple aspects to achieve the breeding programs since they can be used to choose pleasant rice appropriate food quality. Food composition, aroma, flavor, or aromas. In the latter, the feasibility of using a polymer-coated- nutritional properties are among the most important features capillary for the separation of anionic metabolites in both orange that need to be evaluated in food quality assessments. juice and wine has been demonstrated [186]. It offers a com- Several MS-based approaches have been used to de- plementary coverage of the metabolome of these samples to termine food quality. (e most recent publications have those provided by other analytical techniques. mainly used nontargeted MS-based approaches, which very In addition to spray-based ionization techniques [203–206], often included the coupling LC-MS and/or GC-MS. mass spectrometry imaging (MSI) is another useful technique Among LC-MS analytical methods, LC-HRMS tech- for food safety and quality control through monitoring the niques have been used for quality evaluation of raw turmeric spatial distribution of bioactive components and contaminants form different regions [177], for the discrimination of grapes in food samples. Until recently, MSI was largely performed with according to plant sterol content [178], for the analysis of the MALDI. MALDI-TOF-MSI was successfully applied to in- metabolome of the Graciano Vitis vinifera wine variety vestigate the distribution of toxic glycoalkaloids in potato tubers [179], and for the investigation of the quality and authen- [207], to identify the site of capsaicin in Capsicum fruits [208], ticity of saffron [180] and strawberries . Moreover, methods and to observe both the tissue site of 10 anthocyanin species in based on the UPLC combined with HRMS were developed blueberries [209] and the posttranslational modified sites in the to assess the authentication and the evaluation of possible alpha-melanocyte-stimulatin hormone for carp and goldfish adulterations in saffron [181] and fruits juices [182, 183]. (e pituitary tissue, as well as their ratio change under different last two methods rely on the feasibility of the application of environmental conditions [210]. Although MALDI-MSI can the UPLC-QToF platform to perform both nontargeted and detect compounds in a tissue section without extraction, pu- targeted methods to select potential biomarkers, which rification, separation, or labeling, the slow speed of the analysis should make it possible to develop a targeted method (less and the need for matrix deposition in MALDI-MS are critical sophisticated instrumentation and simpler data analysis) for disadvantages in food imaging applications because they in- routine analysis. Similarly, the combination of nontargeted volve analyte diffusion able to affect the original molecular and targeted methods was reported for the qualitative distribution. (e development of various ambient ionization analysis of curcuminoids in turmeric [184]. In this case, techniques revived interest in MSI because of the direct surface nontargeted analysis was performed by using LC-QTOF- sampling in front of a mass spectrometer with submillimeter MS/MS and the targeted approach by LC-QTRAP-MS/MS. resolution and no sample preparation. (ese techniques permit In the LC-MS-based approaches devoted to food quality rapid, direct measurement of compounds present on the assessment, it is noteworthy to highlight the use of hydrophilic condensed sample phase and have become potential analytical interaction chromatography (HILIC), especially in metab- tools for direct profile-imaging analysis in an atmospheric olomics approaches. HILIC columns allow profiling highly pressure environment, thus being particularly useful for food- polar and hydrophilic compounds providing complementary quality control purposes. metabolic information to reversed-phase LC. In spite of several Although the application of these techniques in food caveats associated to HILIC, such as variability in retention analysis is not yet fully established, some examples can be times, low peak efficiency, and long re-equilibration times after found in literature. As an example, ELDI-MS was applied to gradient elution, this methodology has been successfully used obtain the molecular profiling and spatial distribution of to analyze contamination and degradation of infant formulas particular active components in two edible fungi species [211] [185], to separate and detect marine toxins [186], or to identify as well as alpha-solanine and alpha-chaconine in potato [212]. biomarkers of meat quality [187]. DESI-MSI was used to reveal the spatial distribution of GC-MS-based approaches have also been widely used to chlorogenic acids and sucrose across the bean endosperm evaluate food quality. In these approaches, GC was coupled to [213], as well as the spatial and temporal distribution of a huge diversity of mass analyzers: from simple MS in- rohitukine and related compounds during various stages of struments, like quadrupole (Q) [188–192], IT [193], to high- seed development [214]. LAESI allowed macroscopic and resolution instruments [194–198], as well as hybrid analyzers microscopic imaging of pesticides, mycotoxins, and plant [199–201]. Studies of the effect of volatile compounds for the metabolites in different matrices [215]. classification of saffron based on the concentration of biomarkers [188], classification of olive oils on the basis of their 6. Conclusions quality parameters [200], the establishment of differences between wine grape cultivars [194], or the detection of milk or MS-based techniques represent a highly valuable tool for meat adulteration [78, 190] are only some of examples relying environmental, bioanalytical, food safety, and food-quality on the use of GC-MS platforms in food quality analysis. control purposes and their application in these fields Besides the much more common LC-MS and GC-MS strongly increased in the past years. Despite the numerous platforms to assess food quality, comprehensive two- advantages of MS-based methods, one of the most chal- dimensional GC [202] and CE methods [186] coupled to lenging aspects is still related to the analysis of complex TOF analyzers have also been applied. GC × GC allowed the matrices for the detection of nontarget or unknown com- creation of a panel of biomarkers of rice flavor quality through pounds. (e creation of detailed libraries of compounds, establishing associations between volatile metabolites and including MS-based information such as accurate mass, 16 Journal of Analytical Methods in Chemistry isotopic patterns, and collision-induced fragmentation, is OTPAC: Octadecyldimethyl[3-(trimethoxysilyl)propyl] strongly demanded. ammonium chloride Innovations in ambient MS allowed the development of PAHs: Polycyclic aromatic hydrocarbons analytical methods characterized by high-throughput and PB: Paraffin barrier minimal sample preparation, thus allowing the analysis of PBDE: Polybrominated diphenyl ether samples in their native ambient. However, an important PCB: Polychlorobiphenyl feature to be discussed when ambient MS methods are used PDMS: Polydimethylsiloxane is related to the concentration of the detected compounds on PESI: Probe electrospray ionization the sample surface that might not represent the actual PFC: Perfluorinated compound concentration in the whole sample, thus not matching the PMMA: Poly(methyl methacrylate) requirements of current legislation or official methods of PLA: Polylactate analysis. PLE: Pressurized liquid extraction PLLA: Poly-l-lactic acid PSI: Paper spray ionization Nomenclature PTFE: Polytetrafluoroethylene Q: Quadrupole CE: Capillary electrophoresis QqQ: Triple quadrupole CNT: Carbon nanotube QuEChERS: Quick easy cheap effective rugged safe DART: Direct analysis in real time RSD: Relative standard deviation DESI: Desorption electrospray ionization SCWT: Surface-coated wooden tip DMPA: N,N-dimethyl-p-phenylenediamine SFME: Slug-flow microextraction EASI: Easy ambient sonic-spray ionization SLE: Solid-liquid extraction EC: Electrochemical SMB: Supersonic molecular beam EESI: Extractive electrospray ionization SPE: Solid-phase extraction EI: Electron ionization SPME: Solid-phase microextraction ELDI: Electrospray laser desorption ionization TLC: (in-layer chromatography ESI: Electrospray ionization TOF: Time of flight ESTASI: Electrostatic spray ionization TPPB: Tetraphenylphosphonium bromide FA: Fatty acid UPLC: Ultraperformance liquid chromatography FFA: Free fatty acid UTLC: Ultrathin-layer chromatography FI: Field-induced WT: Wooden tip FT-ICR: Fourier-transform ion cyclotron resonance ZV-PSI: Zero volt-paper spray ionization. GC: Gas chromatography GC × GC: Two-dimensional gas chromatography GO: Graphene oxide Conflicts of Interest GOM: Graphene oxide membrane HMF: Hydroxymethylfurfural (e authors declare that they have no conflicts of interest. HILIC: Hydrophilic interaction liquid chromatography References HPLC: High-performance liquid chromatography HRMS: High-resolution mass spectrometry [1] T. Kind and O. Fiehn, “Advances in structure elucidation of LIT: Linear ion trap small molecules using mass spectrometry,” Bioanalytical IT: Ion trap Reviews, vol. 2, no. 1–4, pp. 23–60, 2010. [2] R. Chen, J. Deng, L. Fang et al., “Recent applications of LAESI: Laser ablation electrospray ionization ambient ionization mass spectrometry in environmental LC: Liquid chromatography analysis,” Trends in Environmental Analytical Chemistry, LEI: Liquid-electron ionization vol. 15, pp. 1–11, 2017. LESA: Liquid extraction surface analysis [3] M. Smoluch, P. Mielczarek, and J. Silberring, “Plasma-based LLE: Liquid-liquid extraction ambient ionization mass spectrometry in bioanalytical sci- LS: Liquid sample ences,” Mass Spectrometry Reviews, vol. 35, no. 1, pp. 22–34, LOQ: Limit of quantitation 2016. MALDI: Matrix-assisted laser desorption/ionization [4] X. Ma and Z. Ouyang, “Ambient ionization and miniature ME: Matrix effect mass spectrometry system for chemical and biological MG: Malachite green analysis,” TrAC Trends in Analytical Chemistry, vol. 85, pp. 10–19, 2016. MIP: Molecularly imprinted polymer [5] C. W. Klampfl and M. Himmelsbach, “Direct ionization MRL: Maximum residue limit methods in mass spectrometry: an overview,” Analytica MRM: Multiple reaction monitoring Chimica Acta, vol. 890, pp. 44–59, 2015. MS: Mass spectrometry [6] C. Y. Shi and C. H. Deng, “Recent advances in inorganic MSI: Mass spectrometry imaging materials for LDI-MS analysis of small molecules,” Analyst, MS/MS: Tandem mass spectrometry vol. 141, no. 10, pp. 2816–2826, 2016. Journal of Analytical Methods in Chemistry 17

[7] B. B. Schneider, E. G. Nazarov, F. Londry, P. Vouros, and [23] A. G. Marshall and C. L. Hendrickson, “High-resolution mass T. R. Covey, “Differential mobility spectrometry/mass spectrometers,” Annual Review of Analytical Chemistry, vol. 1, spectrometry history, theory, design optimization, simula- no. 1, pp. 579–599, 2008. tions, and applications,” Mass Spectrometry Reviews, vol. 35, [24] Y. Fu, C. Zhao, X. Lu, and G. Xu, “Nontargeted screening of no. 6, pp. 687–737, 2015. chemical contaminants and illegal additives in food based on [8] T. J. Kauppila, J. A. Syage, and T. Benter, “Recent de- liquid chromatography-high resolution mass spectrometry,” velopments in atmospheric pressure photoionization-mass TrAC Trends in Analytical Chemistry, vol. 96, pp. 89–98, spectrometry,” Mass Spectrometry Reviews, vol. 36, no. 3, 2017. pp. 423–449, 2017. [25] M. Mattarozzi, M. Milioli, C. Cavalieri, F. Bianchi, and [9] P. M. Peacock, W. J. Zhang, and S. Trimpin, “Advances in M. Careri, “Rapid desorption electrospray ionization-high ionization for mass spectrometry,” Analytical Chemistry, resolution mass spectrometry method for the analysis of vol. 89, no. 1, pp. 372–388, 2017. melamine migration from melamine tableware,” Talanta, [10] Z. Takats,` J. M. Wiseman, B. Gologan, and R. G. Cooks, vol. 101, pp. 453–459, 2012. “Mass spectrometry sampling under ambient conditions [26] B. O. Crews, A. J. Pesce, R. West, H. Nguyen, and with desorption electrospray ionization,” Science, vol. 306, R. L. Fitzgerald, “Evaluation of high-resolution mass spec- no. 5695, pp. 471–473, 2004. trometry for urine toxicology screening in a pain manage- [11] H. Wang, J. Liu, R. G. Cooks, and Z. Ouyang, “Paper spray ment setting,” Journal of Analytical Toxicology, vol. 36, no. 9, for direct analysis of complex mixtures using mass spec- pp. 601–607, 2012. trometry,” Angewandte Chemie International Edition, [27] F. Guale, S. Shahreza, J. P. Waltersheid, H. H. Chen, vol. 49, no. 5, pp. 877–880, 2010. C. Arndt, and A. T. Kelly, “Validation of LC-TOF-MS [12] P. Nemes and A. Vertes, “Laser ablation electrospray ioni- screening for drugs, metabolites, and collateral com- zation for atmospheric pressure, in vivo, and imaging mass pounds in forensic toxicology specimens,” Journal of Ana- spectrometry,” Analytical Chemistry, vol. 79, no. 21, lytical Toxicology, vol. 37, no. 1, pp. 17–24, 2013. pp. 8098–8106, 2007. [28] F. W. McLafferty, Interpretation of Mass Spectra, University [13] L. Li and K. A. Schug, “On- and off-line coupling of sepa- Science Books, Mill Valley, CA, USA, 1980. ration techniques to ambient ionization mass spectrometry,” [29] A. Cappiello, G. Famiglini, P. Palma, E. Pierini, V. Termopoli, LCGC North America, vol. 9, no. 4, pp. 8–14, 2011. and H. Trufelli, “Direct-EI in LC-MS: towards a universal [14] A. Herrmann, J. Rosen, D. Jansson, and K. E. Hellenas, detector for small-molecule applications,” Mass Spectrometry “Evaluation of a generic multi-analyte method for detection Reviews, vol. 30, no. 6, pp. 1242–1255, 2011. of >100 representative compounds correlated to emergency [30] P. Palma, G. Famiglini, H. Trufelli, E. Pierini, V. Termopoli, events in 19 food types by ultrahigh-pressure liquid and A. Cappiello, “Electron ionization in LC-MS: recent chromatography-tandem mass spectrometry,” Journal of developments and applications of the direct-EI LC-MS in- Chromatography A, vol. 1235, pp. 115–124, 2012. terface,” Analytical and Bioanalytical Chemistry, vol. 399, [15] F. T. Peters, “Recent advances of liquid chromatography- no. 8, pp. 2683–2693, 2011. (tandem) mass spectrometry in clinical and forensic toxi- [31] A. Cappiello, G. Famiglini, P. Palma, V. Termopoli, cology,” Clinical Biochemistry, vol. 44, no. 1, pp. 54–65 2011. F. Capriotti, and N. Cellar, “Identification potential of direct- [16] D. Remane, D. K. Wissenbach, and F. T. Peters, “Recent EI LC–MS interfacing in small-molecule applications,” advances of liquid chromatography–(tandem) mass spec- Journal of Separation Science, vol. 5, pp. 13–17, 2013. trometry in clinical and forensic toxicology—an update,” [32] V. Termopoli, G. Famiglini, P. Palma, M. Piergiovanni, and Clinical Biochemistry, vol. 49, no. 13-14, pp. 1051–1071, 2016. A. Cappiello, “Atmospheric pressure vaporization mecha- [17] F. T. Peters and D. Remane, “Aspects of matrix effects in nism for coupling a liquid phase with electron ionization applications of liquid chromatography-mass spectrometry to mass spectrometry,” Analytical Chemistry, vol. 89, no. 3, forensic and clinical toxicology–a review,” Analytical and pp. 2049–2056, 2017. Bioanalytical Chemistry, vol. 403, no. 8, pp. 2155–2172, 2012. [33] F. Rigano, A. Albergamo, D. Sciarrone, M. Beccaria, G. Purcaro, [18] F. Gosetti, E. Mazzucco, D. Zampieri, and M. C. Gennaro, and L. Mondello, “Nano liquid chromatography directly cou- “Signal suppression/enhancement in high-performance pled to electron ionization mass spectrometry for free fatty acid liquid chromatography tandem mass spectrometry,” Jour- elucidation in mussel,” Analytical Chemistry, vol. 88, no. 7, nal of Chromatography A, vol. 1217, no. 25, pp. 3929–3937, pp. 4021–4028, 2016. 2010. [34] B. Seemann, T. Alon, S. Tsizin, A. B. Fialkov, and A. Amirav, [19] A. K. Malik, C. Blasco, and Y. Pico, “Liquid chromatography- “Electron ionization LC-MS with supersonic molecular mass spectrometry in food safety,” Journal of Chromatography beams-the new concept, benefits and applications,” Journal A, vol. 1217, no. 25, pp. 4018–4040, 2010. of Mass Spectrometry, vol. 50, no. 11, pp. 1252–1263, 2015. [20] A. Kaufmann, “(e current role of high-resolution mass [35] G. A. Harris, A. S. Galhena, and F. M. Fernandez, “Ambient spectrometry in food analysis,” Analytical and Bioanalytical sampling/ionization mass spectrometry: applications and Chemistry, vol. 403, no. 5, pp. 1233–1249, 2013. current trends,” Analytical Chemistry, vol. 83, no. 12, [21] F. Hernandez, J. V. Sancho, M. Ibanez, E. Abad, T. Portoles, pp. 4508–4538, 2011. and L. Mattioli, “Current use of high-resolution mass [36] M. Domin and R. Cody, Ambient Ionization Mass Spec- spectrometry in the environmental sciences,” Analytical and trometry, Royal Society of Chemistry, Cambridge, UK, 2015. Bioanalytical Chemistry, vol. 403, no. 5, pp. 1251–1264, 2012. [37] L. Magrini, G. Famiglini, P. Palma, V. Termopoli, and [22] M. M. Gomez-Ramos, C. Ferrer, O. Malato, A. Aguera, and A. Cappiello, “Boosting the detection potential of liquid A. R. Fernandez-Alba, “Liquid chromatography-high-resolution chromatography-electron ionization mass spectrometry mass spectrometry for pesticide residue analysis in fruit and using a ceramic coated ion source,” Journal of the American vegetables: screening and quantitative studies,” Journal of Chro- Society for Mass Spectrometry, vol. 27, no. 1, pp. 153–160, matography A, vol. 1287, pp. 24–37, 2013. 2016. 18 Journal of Analytical Methods in Chemistry

[38] N. Riboni, L. Magrini, F. Bianchi, M. Careri, and A. Cappiello, [52] F. Chen, L. Lin, J. Zhang, Z. He, K. Uchiyama, and J. M. Lin, “Sol-gel coated ion sources for liquid chromatography-direct “Single-cell analysis using drop-on-demand inkjet printing electron ionization mass spectrometry,” Analytica Chimica and probe electrospray ionization mass spectrometry,” Acta, vol. 978, pp. 35–41, 2017. Analytical Chemistry, vol. 88, no. 8, pp. 4354–4360, 2016. [39] F. Bianchi, A. Gregori, G. Braun, C. Crescenzi, and [53] X. Gong, Y. Zhao, S. Cai et al., “Single cell analysis with probe M. Careri, “Micro-solid-phase extraction coupled to de- ESI-mass spectrometry: detection of metabolites at cellular sorption electrospray ionization-high-resolution mass and subcellular levels,” Analytical Chemistry, vol. 86, no. 8, spectrometry for the analysis of explosives in soil,” Analytical pp. 3809–3816, 2014. and Bioanalytical Chemistry, vol. 407, no. 3, pp. 931–938, [54] Z. Takats, J. M. Wiseman, D. R. Ifa, and R. G. Cooks, 2014. “Desorption electrospray ionization (DESI) analysis of [40] N. L. Sanders, S. Kothari, G. Huang, G. Salazar, and tryptic digests/peptides,” Cold Spring Harbor Protocols, R. G. Cooks, “Detection of explosives as negative ions di- vol. 2008, no. 5, p. pdb.prot4993, 2008. rectly from surfaces using a miniature mass spectrometer,” [55] S. P. Pasilis, V. Kertesz, G. J. Van Berkel, M. Schulz, and Analytical Chemistry, vol. 82, no. 12, pp. 5313–5316, 2010. S. Schorcht, “Using HPTLC/DESI-MS for peptide identifi- [41] I. Cotte-Rodr´ıguez, Z. Takats, N. Talaty, H. Chen, and cation in 1D separations of tryptic protein digests,” Ana- R. G. Cooks, “Desorption electrospray ionization of explo- lytical and Bioanalytical Chemistry, vol. 391, no. 1, sives on surfaces: sensitivity and selectivity enhancement by pp. 317–324, 2008. reactive desorption electrospray ionization,” Analytical [56] G. Parsiegla, B. Shrestha, F. Carriere, and A. Vertes, “Direct Chemistry, vol. 77, no. 21, pp. 6755–6764, 2005. analysis of phycobilisomal antenna proteins and metabolites [42] N. Talaty, C. C. Mulligan, D. R. Justes, A. U. Jackson, in small cyanobacterial populations by laser ablation elec- R. J. Noll, and R. G. Cooks, “Fabric analysis by ambient mass trospray ionization mass spectrometry,” Analytical Chem- spectrometry for explosives and drugs,” Analyst, vol. 133, istry, vol. 84, no. 1, pp. 34–38, 2012. no. 11, pp. 1532–1540, 2008. [57] Z. P. Yao, “Characterization of proteins by ambient mass [43] I. Cotte-Rodriguez, H. Hernandez-Soto, H. Chen, and spectrometry,” Mass Spectrometry Reviews, vol. 31, no. 4, R. G. Cooks, “In situ trace detection of peroxide explosives by pp. 437–447, 2012. desorption electrospray ionization and desorption atmo- [58] M. Morelato, A. Beavis, P. Kirkbride, and C. Roux, “Forensic applications of desorption electrospray ionisation mass spheric pressure chemical ionization,” Analytical Chemistry, spectrometry (DESI-MS),” Forensic Science International, vol. 80, no. 5, pp. 1512–1519, 2008. vol. 226, no. 1–3, pp. 10–21, 2013. [44] J. M. Wiseman, D. R. Ifa, Y. Zhu et al., “Desorption elec- [59] C. Ibáñez,V. Garc´ıa-Cañas, A. Valdés,and C. Simó,“Novel trospray ionization mass spectrometry: imaging drugs and MS-based approaches and applications in food metab- metabolites in tissues,” Proceedings of the National Academy olomics,” TrAC Trends in Analytical Chemistry, vol. 52, of Sciences, vol. 105, no. 47, pp. 18120–18125, 2008. pp. 100–111, 2013. [45] T. J. Kauppila, N. Talaty, T. Kuuranne, T. Kotiaho, R. Kostiainen, [60] M. Castro-Puyana and M. Herrero, “Metabolomics ap- and R. G. Cooks, “Rapid analysis of metabolites and drugs of proaches based on mass spectrometry for food safety, quality abuse from urine samples by desorption electrospray ionization- and traceability,” TrAC Trends in Analytical Chemistry, mass spectrometry,” Analyst, vol. 132, no. 9, pp. 868–875, 2007. vol. 52, pp. 74–87, 2013. [46] T. J. Kauppila, J. M. Wiseman, R. A. Ketola, T. Kotiaho, [61] P. Nemes and A. Vertes, “Ambient mass spectrometry for in R. G. Cooks, and R. Kostiainen, “Desorption electrospray vivo local analysis and in situ molecular tissue imaging,” ionization mass spectrometry for the analysis of pharma- TrAC Trends in Analytical Chemistry, vol. 34, pp. 22–34, ceuticals and metabolites,” Rapid Communications in Mass 2012. Spectrometry, vol. 20, no. 3, pp. 387–392, 2006. [62] E. R. St John, M. Rossi, P. Pruski, A. Darzi, and Z. Takats, [47] Z. Lin, S. Zhang, M. Zhao, C. Yang, D. Chen, and X. Zhang, “Intraoperative tissue identification by mass spectrometric “Rapid screening of clenbuterol in urine samples by de- technologies,” TrAC Trends in Analytical Chemistry, vol. 85, sorption electrospray ionization tandem mass spectrome- pp. 2–9, 2016. try,” Rapid Communications in Mass Spectrometry, vol. 22, [63] Y. Yang, Y. Huang, J. Wu, N. Liu, J. Deng, and T. Luan, no. 12, pp. 1882–1888, 2008. “Single-cell analysis by ambient mass spectrometry,” TrAC [48] H. Chen, J. Zheng, X. Zhang, M. Luo, Z. Wang, and X. Qiao, Trends in Analytical Chemistry, vol. 90, pp. 14–26, 2017. “Surface desorption atmospheric pressure chemical ioniza- [64] A. A. Lubin, D. Cabooter, P. Augustijns, and F. Cuyckens, tion mass spectrometry for direct ambient sample analysis “One drop chemical derivatization – DESI-MS analysis for without toxic chemical contamination,” Journal of Mass metabolite structure identification,” Journal of Mass Spec- Spectrometry, vol. 42, no. 8, pp. 1045–1056, 2007. trometry, vol. 50, no. 7, pp. 871–878, 2015. [49] R. J. Fussell, D. Chan, and M. Sharman, “An assessment of [65] T. A. Brown, H. Chen, and R. N. Zare, “Identification of atmospheric-pressure solids-analysis probes for the de- fleeting electrochemical reaction intermediates using detection of chemicals in food,” TrAC Trends in Analytical sorption electrospray ionization mass spectrometry,” Journal Chemistry, vol. 29, no. 11, pp. 1326–1335, 2010. of the American Chemical Society, vol. 137, no. 23, [50] C. Black, O. P. Chevallier, and C. T. Elliott, “(e current and pp. 7274–7277, 2015. potential applications of ambient mass spectrometry in [66] W. D. Looi, B. Brown, L. Chamand, and A. Brajter-Toth, detecting food fraud,” TrAC Trends in Analytical Chemistry, “Merits of online electrochemistry liquid sample desorption vol. 82, pp. 268–278, 2016. electrospray ionization mass spectrometry (EC/LS DESI [51] E. Hiyama, A. Ali, S. Amer et al., “Direct lipido-metabolomics MS),” Analytical and Bioanalytical Chemistry, vol. 408, no. 9, of single floating cells for analysis of circulating tumor cells by pp. 2227–2238, 2016. live single-cell mass spectrometry,” Analytical Sciences, vol. 31, [67] Y. Cai, D. Adams, and H. Chen, “A new splitting method for no. 12, pp. 1215–1517, 2015. both analytical and preparative LC/MS,” Journal of the Journal of Analytical Methods in Chemistry 19

American Society for Mass Spectrometry, vol. 25, no. 2, Environmental Science & Technology, vol. 47, no. 13, pp. 286–292, 2014. pp. 7324–7331, 2013. [68] Y. Cai, Y. Liu, R. Helmy, and H. Chen, “Coupling of ultrafast [82] G. K. Koyanagi, V. Blagojevic, and D. K. Bohme, “Applications LC with mass spectrometry by DESI,” Journal of Ce of extractive electrospray ionization (EESI) in analytical American Society for Mass Spectrometry, vol. 25, no. 10, chemistry,” International Journal of Mass Spectrometry, pp. 1820–1823, 2014. vol. 379, pp. 146–150, 2015. [69] Y. Ren, M. N. McLuckey, J. Liu, and Z. Ouyang, “Direct mass [83] M. Deng, T. Yu, H. Luo, T. Zhu, X. Huang, and L. Luo, spectrometry analysis of biofluid samples using slug-flow “Direct detection of multiple pesticides in honey by neutral microextraction nano-electrospray ionization,” Angewandte desorption-extractive electrospray ionization mass spec- Chemie International Edition, vol. 53, no. 51, pp. 14124– trometry,” International Journal of Mass Spectrometry, 14127, 2014. vol. 422, pp. 111–118, 2017. [70] J. Cain, A. Laskin, M. R. Kholghy, M. J. (omson, and [84] N. Xu, Z. Q. Zhu, S. P. Yang et al., “Direct detection of amino H. Wang, “Molecular characterization of organic content of acids using extractive electrospray ionization tandem mass soot along the centerline of a coflow diffusion flame,” spectrometry,” Chinese Journal of Analytical Chemistry, Physical Chemistry Chemical Physics, vol. 16, no. 47, vol. 41, no. 4, p. 523, 2013. pp. 25862–25875, 2014. [85] X. Li, X. Fang, Z. Yu et al., “Direct quantification of cre- [71] S. Tao, X. Lu, N. Levac et al., “Molecular characterization of atinine in human urine by using isotope dilution extractive organosulfates in organic aerosols from Shanghai and Los electrospray ionization tandem mass spectrometry,” Ana- Angeles urban areas by nanospray-desorption electrospray ionization high-resolution mass spectrometry,” Environ- lytica Chimica Acta, vol. 748, pp. 53–57, 2012. [86] R. S. Jacobson, R. L. (urston, B. Shrestha, and A. Vertes, “In mental Science & Technology, vol. 48, no. 18, pp. 10993– 11001, 2014. situ analysis of small populations of adherent mammalian [72] E. J. Boone, A. Laskin, J. Laskin et al., “Aqueous processing of cells using laser ablation electrospray ionization mass atmospheric organic particles in cloud water collected via spectrometry in transmission geometry,” Analytica Chimica aircraft sampling,” Environmental Science & Technology, Acta, vol. 87, no. 24, pp. 12130–12136, 2015. vol. 49, no. 14, pp. 8523–8530, 2015. [87] L. R. Compton, B. Reschke, J. Friend, M. Powell, and [73] C. Cardoso-Palacios and I. Lanekoff, “Direct analysis of A. Vertes, “Remote laser ablation electrospray ionization pharmaceutical drugs using nano-DESI MS,” Journal of mass spectrometry for non-proximate analysis of biological Analytical Methods in Chemistry, vol. 2016, Article ID tissues,” Rapid Communications in Mass Spectrometry, 3591908, 6 pages, 2016. vol. 29, no. 1, pp. 67–73, 2015. [74] J. Watrous, P. Roach, B. Heath, T. Alexandrov, J. Laskin, and [88] L. C. Duarte, T. C. de Carvalho, E. O. Lobo-Júnior, P. V. Abdelnur, P. C. Dorrestein, “Metabolic profiling directly from the Petri B. G. Vaza, and W. K. T. Coltro, “3D printing of microfluidic dish using nanospray desorption electrospray ionization devices for paper-assisted direct spray ionization mass spec- imaging mass spectrometry,” Analytical Chemistry, vol. 85, trometry,” Analytical Methods, vol. 8, no. 3, pp. 496–503, 2016. no. 21, pp. 10385–10391, 2013. [89] G. I. J. Salentijn, H. P. Permentier, and E. Verpoorte, “3D- [75] H. M. Bergman and I. Lanekoff, “Profiling and quantifying printed paper spray ionization cartridge with fast wetting endogenous molecules in single cells using nano-DESI MS,” and continuous solvent supply features,” Analytical Chem- Analyst, vol. 142, no. 19, pp. 3639–3647, 2017. istry, vol. 86, no. 23, pp. 11657–11665, 2014. [76] H. M. Bergman, E. Lundin, M. Andersson, and I. Lanekoff, [90] L. Shen, J. Zhang, Q. Yang, N. E. Manicke, and Z. Ouyang, “Quantitative mass spectrometry imaging of small-molecule “High throughput paper spray mass spectrometry analysis,” neurotransmitters in rat brain tissue sections using nano- Clinica Chimica Acta, vol. 420, pp. 28–33, 2013. spray desorption electrospray ionization,” Analyst, vol. 141, [91] J. Deng, W. Wang, Y. Yang et al., “Slug-flow microextraction no. 12, pp. 3686–3695, 2016. coupled with paper spray mass spectrometry for rapid [77] C. C. Hsu, P. T. Chou, and R. N. Zare, “Imaging of proteins analysis of complex samples,” Analytica Chimica Acta, in tissue samples using nanospray desorption electrospray vol. 940, pp. 143–149, 2016. ionization mass spectrometry,” Analytical Chemistry, vol. 87, [92] C. Zhang and N. E. Manicke, “Development of a paper spray no. 22, pp. 11171–11175, 2015. mass spectrometry cartridge with integrated solid phase [78] D. Duncan, H. M. Bergman, and I. Lanekoff, “A pneu- extraction for bioanalysis,” Analytical Chemistry, vol. 87, matically assisted nanospray desorption electrospray ioni- no. 12, pp. 6212–6219, 2015. zation source for increased solvent versatility and enhanced [93] B. Yang, F. Wang, W. Deng et al., “Wooden-tip electrospray metabolite detection from tissue,” Analyst, vol. 142, no. 18, ionization mass spectrometry for trace analysis of toxic and pp. 3424–3431, 2017. [79] H. W. Chen, A. Venter, and R. G. Cooks, “Extractive elec- hazardous compounds in food samples,” Analytical trospray ionization for direct analysis of undiluted urine, milk Methods, vol. 7, no. 14, pp. 5886–5890, 2015. and other complex mixtures without sample preparation,” [94] G. Z. Xin, B. Hu, Z. Q. Shi et al., “Rapid identification of plant Chemical Communications, vol. 42, no. 19, pp. 2042–2044, materials by wooden-tip electrospray ionization mass 2006. spectrometry and a strategy to differentiate the bulbs of [80] R. Wang, A. J. Gröhn,L. Zhu et al., “On the mechanism of Fritillaria,” Analytica Chimica Acta, vol. 820, pp. 84–91, extractive electrospray ionization (EESI) in the dual-spray 2014. configuration,” Analytical and Bioanalytical Chemistry, [95] Y. Yang and J. Deng, “Internal standard mass spectrum vol. 402, no. 8, pp. 2633–2643, 2012. fingerprint: a novel strategy for rapid assessing the quality of [81] P. J. Gallimore and M. Kalberer, “Characterizing an ex- Shuang-Huang-Lian oral liquid using wooden-tip electro- tractive electrospray ionization (EESI) source for the online spray ionization mass spectrometry,” Analytica Chimica mass spectrometry analysis of organic aerosols,” Acta, vol. 837, pp. 83–92, 2014. 20 Journal of Analytical Methods in Chemistry

[96] H. K. Chen, C. H. Lin, J. T. Liu, and C. H. Lin, “Electrospray [110] K. J. Ewing, D. Gibson, J. Sanghera, and F. Miklos, “Desorption ionization using a bamboo pen nib,” International Journal of electrospray ionization–mass spectrometric analysis of low Mass Spectrometry, vol. 356, pp. 37–40, 2013. vapor pressure chemical particulates collected from a surface,” [97] B. Hu and Z. P. Yao, “Detection of native proteins using Analytica Chimica Acta, vol. 853, pp. 368–374, 2015. solid-substrate electrospray ionization mass spectrometry [111] R. G. Hemalatha, M. A. Ganayee, and T. Pradeep, “Elec- with nonpolar solvents,” Analytica Chimica Acta, vol. 1004, trospun nanofiber mats as “smart surfaces” for desorption pp. 51–57, 2017. electrospray ionization mass spectrometry (DESI MS)-based [98] Y. Yang, J. Deng, and Z. P. Yao, “Field-induced wooden-tip analysis and imprint imaging,” Analytical Chemistry, vol. 88, electrospray ionization mass spectrometry for high- no. 11, pp. 5710–5717, 2016. throughput analysis of herbal medicines,” Analytica Chi- [112] C. H. Lin, W. C. Liao, H. K. Chen, and T. Y. Kuo, “Paper mica Acta, vol. 887, pp. 127–137, 2015. spray-MS for bioanalysis,” Bioanalysis, vol. 6, no. 2, pp. 1–10, [99] Y. W. Liou, K. Y. Chang, and C. H. Lin, “Sampling and 2014. profiling caffeine and its metabolites from an eyelid using [113] Q. Yang, H. Wang, J. D. Maas et al., “Paper spray ionization a watercolor pen based on electrospray ionization/mass devices for direct, biomedical analysis using mass spec- spectrometry,” International Journal of Mass Spectrometry, trometry,” International Journal of Mass Spectrometry, vol. 422, pp. 51–55, 2017. vol. 312, pp. 201–207, 2012. [100] N. Pan, W. Rao, N. R. Kothapalli, R. Liu, A. W. G. Burgett, [114] H. Evard, A. Kruve, R. Lõhmus, and I. Leito, “Paper spray and Z. Yang, “(e single-probe: a miniaturized multifunc- ionization mass spectrometry: study of a method for fast- tional device for single cell mass spectrometry analysis,” screening analysis of pesticides in fruits and vegetables,” Analytical Chemistry, vol. 86, no. 19, pp. 9376–9380, 2014. Journal of Food Composition and Analysis, vol. 41, pp. 221– [101] C. Shiea, Y. L. Huang, S. C. Cheng, Y. L. Chen, and J. Shiea, 225, 2015. “Determination of elemental composition of metals using [115] Z. P. Zhang, X. N. Liu, and Y. J. Zheng, “Ambient ionization- ambient organic mass spectrometry,” Analytica Chimica paper spray ionization and its application,” Chinese Journal Acta, vol. 968, pp. 50–57, 2017. of Analytical Chemistry, vol. 42, no. 1, pp. 145–152, 2014. [102] L. Qiao, R. Sartor, N. Gasilova et al., “Electrostatic-spray [116] P. H. Lai, P. C. Chen, Y. W. Liao, J. T. Liu, C. C. Chen, and ionization mass spectrometry,” Analytical Chemistry, vol. 84, C. H. Lin, “Comparison of gampi paper and nanofibers to no. 17, pp. 7422–7430, 2012. chromatography paper used in paper spray-mass spec- [103] Z. Takats, J. M. Wiseman, and R. G. Cooks, “Ambient mass trometry,” International Journal of Mass Spectrometry, spectrometry using desorption electrospray ionization vol. 375, pp. 14–17, 2015. (DESI): instrumentation, mechanisms and applications in [117] T. C. Colletes, P. T. Garcia, R. B. Campanha et al., “A new forensics, chemistry, and biology,” Journal of Mass Spec- insert sample approach to paper spray mass spectrometry: trometry, vol. 40, no. 10, pp. 1261–1275, 2005. a paper substrate with paraffin barriers,” Analyst, vol. 141, [104] A. Penna, M. Careri, N. D. Spencer, and A. Rossi, “Effects of no. 5, pp. 1707–1713, 2016. tailored surface chemistry on desorption electrospray ioni- [118] Z. Zhang, W. Xu, N. E. Manicke, R. G. Cooks, and zation mass spectrometry: a surface-analytical study by XPS Z. Ouyang, “Silica coated paper substrate for paper-spray and AFM,” Journal of the American Society for Mass Spec- analysis of therapeutic drugs in dried blood spots,” Ana- trometry, vol. 26, no. 8, pp. 1311–1319, 2015. lytical Chemistry, vol. 84, no. 2, pp. 931–938, 2012. [105] L. Elviri, R. Foresti, A. Bianchera, M. Silvestri, and R. Bettini, [119] R. Narayanan, D. Sarkar, R. G. Cooks, and T. Pradeep, “3D-printed polylactic acid supports for enhanced ionization “Molecular ionization from carbon nanotube paper,” efficiency in desorption electrospray mass spectrometry Angewandte Chemie International Edition, vol. 53, no. 23, analysis of liquid and gel samples,” Talanta, vol. 155, pp. 5936–5940, 2014. pp. 321–328, 2016. [120] S. C. Wei, S. Fan, C. W. Lien et al., “Graphene oxide [106] M. Montowska, W. Rao, M. R. Alexander, G. A. Tucker, and membrane as an efficient extraction and ionization substrate D. A. Barrett, “Tryptic digestion coupled with ambient de- for spray-mass spectrometric analysis of malachite green and sorption electrospray ionization and liquid extraction sur- its metabolite in fish samples,” Analytica Chimica Acta, face analysis mass spectrometry enabling identification of vol. 1003, pp. 42–48, 2018. skeletal muscle proteins in mixtures and distinguishing [121] J. Liu, Y. He, S. Chen, M. Ma, S. Yao, and B. Chen, “New between beef, pork, horse, chicken, and turkey meat,” An- urea-modified paper substrate for enhanced analytical per- alytical Chemistry, vol. 86, no. 9, pp. 4479–4487, 2014. formance of negative ion mode paper spray mass spec- [107] M. T. Dulay, L. S. Eberlin, and R. N. Zare, “Protein analysis trometry,” Talanta, vol. 166, pp. 306–314, 2017. by ambient ionization mass spectrometry using trypsin- [122] I. Pereira, M. F. Rodrigues, A. R. Chaves, and B. G. Vaz, immobilized organosiloxane polymer surfaces,” Analytical “Molecularly imprinted polymer (MIP) membrane assisted Chemistry, vol. 87, no. 24, pp. 12324–12330, 2015. direct spray ionization mass spectrometry for agrochemicals [108] S. Cheng, J. Wang, Y. Cai, J. A. Loo, and H. Chen, “En- screening in foodstuffs,” Talanta, vol. 178, pp. 507–514, 2018. hancing performance of liquid sample desorption electro- [123] B. J. Bills and N. E. Manicke, “Development of a prototype spray ionization mass spectrometry using trap and capillary blood fractionation cartridge for plasma analysis by paper columns,” International Journal of Mass Spectrometry, spray mass spectrometry,” Clinical Mass Spectrometry, vol. 2, vol. 392, pp. 73–79, 2015. pp. 18–24, 2016. [109] S. S. Kanyal, T. T. Habe,¨ C. V. Cushman et al., “Micro- [124] M. Wleklinski, Y. Li, S. Bag et al., “Zero volt paper spray fabrication, separations, and detection by mass spectrometry ionization and its mechanism,” Analytical Chemistry, vol. 87, on ultrathin-layer chromatography plates prepared via the no. 13, pp. 6786–6793, 2015. low-pressure chemical vapor deposition of silicon nitride [125] J. Deng, Y. Yang, L. Fang, L. Lin, H. Zhou, and T. Luan, onto carbon nanotube templates,” Journal of Chromatog- “Coupling solid-phase microextraction with ambient mass raphy A, vol. 1404, pp. 115–123, 2015. spectrometry using surface coated wooden-tip probe for Journal of Analytical Methods in Chemistry 21

rapid analysis of ultra trace perfluorinated compounds in [140] S. S. Shida, S. Nemoto, and R. Matsuda, “Simultaneous complex samples,” Analytical Chemistry, vol. 86, no. 22, determination of acidic pesticides in vegetables and fruits by pp. 11159–11166, 2014. liquid chromatography–tandem mass spectrometry,” Jour- [126] J. Deng, T. Yu, Y. Yao et al., “Surface-coated wooden-tip nal of Environmental Science and Health, Part B, vol. 50, electrospray ionization mass spectrometry for determination no. 3, pp. 151–162, 2015. of trace fluoroquinolone and macrolide antibiotics in water,” [141] C. Rasche, B. Fournes, U. Dirks, and K. Speer, “Multi-residue Analytica Chimica Acta, vol. 954, pp. 52–59, 2017. pesticide analysis (gas chromatography–tandem mass [127] Y. Huang, Y. Ma, H. Hu et al., “Rapid and sensitive detection spectrometry detection)–improvement of the quick, easy, of trace malachite green and its metabolite in aquatic cheap, effective, rugged, and safe method for dried fruits and products using molecularly imprinted polymer-coated fat-rich cereals–benefit and limit of a standardized apple wooden-tip electrospray ionization mass spectrometry,” puréecalibration (screening),” Journal of Chromatography RSC Advances, vol. 7, no. 82, pp. 52091–52100, 2017. A, vol. 1403, pp. 21–31, 2015. [128] B. Hu, P. K. So, and Z. P. Yao, “Electrospray ionization with [142] L. Han, J. Matarrita, Y. Sapozhnikova, and S. J. Lehotay, aluminum foil: a versatile mass spectrometric technique,” “Evaluation of a recent product to remove lipids and other Analytica Chimica Acta, vol. 817, pp. 1–8, 2014. matrix co-extractives in the analysis of pesticide residues and [129] X. Zhong, L. Qiao, B. Liu, and H. H. Girault, “Ambient in situ environmental contaminants in foods,” Journal of Chro- analysis and imaging of both hydrophilic and hydrophobic matography A, vol. 1449, pp. 17–29, 2016. thin layer chromatography plates by electrostatic spray [143] B. D. Morris and R. B. Schriner, “Development of an auto- ionization mass spectrometry,” RSC Advances, vol. 5, no. 92, mated column solid-phase extraction cleanup of QuEChERS pp. 75395–75402, 2015. extracts, using a zirconia-based sorbent, for pesticide residue [130] V. Garcia-Canas, M. Herrero, E. Ibanez, and A. Cifuentes, analyses by LC-MS/MS,” Journal of Agricultural and Food “Food analysis: present, future, and foodomics,” Analytical Chemistry, vol. 63, no. 21, pp. 5107–5119, 2015. Chemistry, vol. 84, no. 23, pp. 10150–10159, 2012. [144] G. Ramadan, M. Al Jabir, N. Alabdulmalik, and A. Mohammed, [131] M. Herrero, C. Simo, V. Garcia-Canas, E. Ibanez, and “Validation of a method for the determination of 120 pesticide A. Cifuentes, “Foodomics: MS-based strategies in modern residues in apples and cucumbers by LC-MS/MS,” Drug Testing food science and nutrition,” Mass Spectrometry Reviews, and Analysis, vol. 8, no. 5-6, pp. 498–510, 2016. vol. 31, no. 1, pp. 49–69, 2012. [145] M. A. Zhao, Y. N. Feng, Y. Z. Zhu, and J. H. Kim, “Multi- [132] F. Cacciola, P. Donato, M. Beccaria, P. Dugo, and residue method for determination of 238 pesticides in Chinese cabbage and cucumber by liquid chromatography- L. Mondello, “Advances in LC-MS for food analysis,” LC GC tandem mass spectrometry: comparison of different purifi- Europe, vol. 25, no. 5, pp. 15–24, 2012. cation procedures,” Journal of Agricultural and Food [133] European Union Commission Decision 2002/657/EC, 2002. Chemistry, vol. 62, no. 47, pp. 11449–11456, 2014. [134] G. C. R. M. Andrade, S. H. Monteiro, J. G. Francisco, [146] N. Fleury-Filho, C. A. Nascimento, E. O. Faria, A. R. Crunivel, L. A. Figueiredo, R. G. Botelho, and V. L. Tornisielo, “Liquid and J. M. Oliveira, “Within laboratory validation of a multi- chromatography–electrospray ionization tandem mass spec- residue method for the analysis of 98 pesticides in mango by LC trometry and dynamic multiple reaction monitoring method for tandem MS,” Food Additives & Contaminants: Part A, vol. 29, determining multiple pesticide residues in tomato,” Food no. 4, pp. 641–656, 2012. Chemistry, vol. 175, pp. 57–65, 2015. [147] T. Cajka, C. Sandy, V. Bachavolva et al., “Streamlining [135] O. Golge and B. Kabak, “Evaluation of QuEChERS sample sample preparation and GC tandem MS analysis of multiple preparation and liquid chromatography–triple-quadrupole pesticide residues in tea,” Analytica Chimica Acta, vol. 743, mass spectrometry method for the determination of 109 pp. 51–60, 2012. pesticide residues in tomatoes,” Food Chemistry, vol. 176, [148] V. Havolt, S. Goscinny, and M. Deridder, “A simple multi- pp. 319–332, 2015. residue method for the determination of pesticides in fruits [136] F. Diniz Madureira, F. A. da Silva Oliveira, W. R. de Souza, and vegetables using a methanolic extraction and ultra-high- A. P. Pontelo, M. L. Gonçalves de Oliveira, and G. Silva, “A performance liquid chromatography-tandem mass spec- multi-residue method for the determination of 90 pesticides trometry: optimization and extension of scope,” Journal of in matrices with a high water content by LC-MS/MS without Chromatography A, vol. 1384, pp. 53–66, 2015. clean-up,” Food Additives & Contaminants: Part A, vol. 29, [149] S. Walorczyk and D. Drozdzynski, “Improvement and ex- no. 4, pp. 665–678, 2012. tension to new analytes of a multi-residue method for the [137] A. Garrido-Frenich, M. M. Mart´ın Fernández, L. D´ıaz determination of pesticides in cereals and dry animal feed Moreno, J. L. Mart´ınez-Vidal, and N. Lopez-Guti´ errez,´ using gas chromatography-tandem quadrupole mass spec- “Multiresidue pesticide analysis of tuber and root com- trometry revisited,” Journal of Chromatography A, vol. 1251, modities by QuEChERS extraction and UPLC coupled to pp. 219–231, 2012. tandem MS,” Journal of AOAC International, vol. 95, no. 5, [150] O. Lacina, M. Zachariasova, J. Urbavolva, M. Vaclavikova, pp. 1319–1330, 2012. T. Cajka, and J. Hajslova, “Critical assessment of extraction [138] O. Golge and B. Kabak, “Determination of 115 pesticide residues methods for the simultaneous determination of pesticide in oranges by high-performance liquid chromatography–triple- residues and mycotoxins in fruits, cereals, spices and oil quadrupole mass spectrometry in combination with QuEChERS seeds employing UPLC tandem MS,” Journal of Chroma- method,” Journal of Food Composition and Analysis, vol. 41, tography A, vol. 1262, pp. 8–18, 2012. pp. 86–97, 2015. [151] J. Wang, W. Chow, and W. Cheung, “Application of a tan- [139] J. Cho, J. Lee, C. U. Lim, and J. Ahn, “Quantification of dem mass spectrometer and core-shell particle column for pesticides in food crops using QuEChERS approaches and the determination of 151 pesticides in grains,” Journal of GCeMS/MS,” Food Additives & Contaminants: Part A, Agricultural and Food Chemistry, vol. 59, no. 16, pp. 8589– vol. 33, no. 12, pp. 1803–1816, 2016. 8608, 2011. 22 Journal of Analytical Methods in Chemistry

[152] Z. He, L. Wang, Y. Peng, M. Luo, W. Wang, and X. Liu, [164] Y. Rodriguez-Carrasco, M. Fattore, S. Albrizio, H. Berrada, “Multiresidue analysis of over 200 pesticides in cereals using and J. Manes, “Occurrence of Fusarium mycotoxins and their a QuEChERS and gas chromatography-tandem mass dietary intake through beer consumption by the European spectrometry-based method,” Food Chemistry, vol. 169, population,” Food Chemistry, vol. 178, pp. 149–155, 2015. pp. 372–380, 2015. [165] C. Juan, J. Mañes, A. Raiola, and A. Ritieni, “Evaluation of [153] A. Palen´ıkova, G. Mart´ınez-Dom´ınguez, F. J. Arrebola, beauvericin and enniatins in Italian cereal products and R. Romero-Gonzalez, S. Hrouzkova, and A. Garrido Frenich, multicereal food by liquid chromatography coupled to triple “Multifamily determination of pesticide residues in soya- quadrupole mass spectrometry,” Food Chemistry, vol. 140, based nutraceutical products by GC/MS-MS,” Food Chem- no. 4, pp. 755–762, 2013. istry, vol. 173, pp. 796–807, 2015. [166] K. Zhang, J. W. Wong, P. Yang et al., “Protocol for an [154] S. Chawla, H. K. Patel, K. M. Vaghela et al., “Development electrospray ionization tandem mass spectral product ion and validation of multi residue analytical method in cotton library: development and application for identification of and groundnut oil for 87 pesticides using low temperature 240 pesticides in foods,” Analytical Chemistry, vol. 84, no. 13, and dispersive cleanup on gas chromatography and liquid pp. 5677–5684, 2012. chromatography-tandem mass spectrometry,” Analytical [167] M. I. Cervera, T. Portoles, E. Pitarch, J. Beltran, and and Bioanalytical Chemistry, vol. 408, no. 3, pp. 983–997, F. Hernandez, “Application of gas chromatography time-of- 2016. flight mass spectrometry for target and non-target analysis of [155] D. L. Christodoulou, P. Kanari, P. Hadjiloizou, and P. Constantivolu, pesticide residues in fruits and vegetables,” Journal of “Pesticide residues analysis in wine by liquid chromatography– Chromatography A, vol. 1244, pp. 168–177, 2012. tandem mass spectrometry and using ethyl acetate extraction [168] F. Lambertini, V. Di Lallo, D. Catellani, M. Mattarozzi, method: validation and pilot survey in real samples,” Journal of Wine M. Careri, and M. Suman, “Reliable liquid chromatography- Research, vol. 26, no. 2, pp. 81–98, 2015. mass spectrometry method for investigation of primary [156] G. R. Chang, H. S. Chen, and F. Y. Lin, “Analysis of banned aromatic amines migration from food packaging and during veterinary drugs and herbicide residues in shellfish by liquid industrial curing of multilayer plastic laminates,” Journal of chromatography-tandem mass spectrometry (LC/MS/MS) and Mass Spectrometry, vol. 49, no. 9, pp. 870–877, 2014. gas chromatography-tandem mass spectrometry (GC/MS/MS),” [169] C. Planche, J. Ratel, F. Mercier, P. Blinet, L. Debrauwer, and Marine Pollution Bulletin, vol. 113, no. 1-2, pp. 579–584, 2016. E. Engel, “Assessment of comprehensive two-dimensional [157] M. E. Dasenaki, C. S. Michali, and N. S. (omaidis, “Analysis gas chromatography-time-of-flight mass spectrometry based of 76 veterinary pharmaceuticals from 13 classes including methods for investigating 206 dioxin-like micropollutants in aminoglycosides in bovine muscle by hydrophilic interaction animal-derived food matrices,” Journal of Chromatography liquid chromatography–tandem mass spectrometry,” Jour- A, vol. 1392, pp. 74–81, 2015. nal of Chromatography A, vol. 1452, pp. 67–80, 2016. [170] X. Wang, P. Li, W. Zhang et al., “Screening for pesticide [158] D. Chen, J. Yu, Y. Tao et al., “Qualitative screening of residues in oil seeds using solid-phase dispersion extraction veterinary anti-microbial agents in tissues, milk, and eggs of and comprehensive two-dimensional gas chromatography food-producing animals using liquid chromatography time-of-flight mass spectrometry,” Journal of Separation coupled with tandem mass spectrometry,” Journal of Science, vol. 35, no. 13, pp. 1634–1643, 2012. Chromatography B, vol. 1017-1018, pp. 82–88, 2016. [171] K. Kalachova, J. Pulkrabova, T. Cajka, L. Drabova, and [159] M. Aznar, A. Rodriguez-Lafuente, P. Alfaro, and C. Nerin, J. Hajslova, “Implementation of comprehensive two- “UPLC-Q-TOF-MS analysis of non-volatile migrants from dimensional GC-time-of-flight-MS for the simultaneous new active packaging materials,” Analytical and Bio- determination of halogenated contaminants and polycyclic analytical Chemistry, vol. 404, no. 6-7, pp. 1945–1957, 2012. aromatic hydrocarbons in fish,” Analytical and Bioanalytical [160] B. Skribic, J. Zivancev, and M. Godula, “Multimycotoxin Chemistry, vol. 403, no. 10, pp. 2813–2824, 2012. analysis of crude extracts of nuts with ultra-high perfor- [172] M. L. Gomez-Perez, P. Plaza-Bolavols, R. Romero-Gonzalez, mance liquid chromatography/tandem mass spectrometry,” J. L. Martinez-Vidal, and A. Garrido-Frenich, “Compre- Journal of Food Composition and Analysis, vol. 34, no. 2, hensive qualitative and quantitative determination of pes- pp. 171–177, 2014. ticides and veterinary drugs in honey using liquid [161] M. Ludovici, C. Ialongo, M. Reverberi, M. Beccaccioli, chromatography-Orbitrap high resolution mass spectrom- M. Scarpari, and V. Scala, “Quantitative profiling of oxy- etry,” Journal of Chromatography A, vol. 1248, pp. 130–138, lipins through comprehensive LC-MS/MS analysis of 2012. Fusarium verticillioides and maize kernels,” Food Additives & [173] P. Perez-Ortega, F. J. Lara-Ortega, J. F. Garc´ıa-Reyes, B. Gilbert- Contaminants: Part A, vol. 31, no. 12, pp. 2026–2033, 2014. Lopez, M. Trojavolwicz, and A. Molina-D´ıaz, “A feasibility study [162] M. Garc´ıa-Altares, A. Casavolva, V. Bane, J. Diogene, of UHPLC-HRMS accurate-mass screening methods for mul- A. Furey, and P. de la Iglesia, “Confirmation of pinnatoxins ticlass testing of organic contaminants in food,” Talanta, and spirolides in shellfish and passive samplers from Cat- vol. 160, pp. 704–712, 2016. alonia (Spain) by liquid chromatography coupled with triple [174] H. G. J. Mol, P. Zomer, and M. de Koning, “Qualitative quadrupole and high-resolution hybrid tandem mass aspects and validation of a screening method for pesticides in spectrometry,” Marine Drugs, vol. 12, no. 6, pp. 3706–3732, vegetables and fruits based on liquid chromatography 2014. coupled to full scan high resolution (Orbitrap) mass spec- [163] Y. Rodriguez-Carrasco, J. Manes, H. Berrada, and C. Juan, trometry,” Analytical and Bioanalytical Chemistry, vol. 403, “Development and validation of a LC-ESI-MS/MS method no. 10, pp. 2891–2908, 2012. for the determination of alternaria toxins alternariol, [175] J. Rubert, K. J. James, J. Manes, and C. Soler, “Applicability of alternariol methyl-ether and tentoxin in tomato and tomato hybrid linear ion trap-high resolution mass spectrometry based products,” Toxins, vol. 8, no. 11, p. 328, 2016. and quadrupole-linear ion trap-mass spectrometry for Journal of Analytical Methods in Chemistry 23

mycotoxin analysis in baby food,” Journal of Chromatog- of commercial balsamic vinegars using gas chromatography- raphy A, vol. 1223, pp. 84–92, 2012. mass spectrometry,” Metabolites, vol. 6, no. 3, p. 22, 2016. [176] D. G. Beach, C. M. Walsh, and P. McCarron, “High- [190] P. Scano, A. Murgia, F. M. Pirisi, and P. Caboni, “A gas throughput quantitative analysis of domoic acid directly chromatography-mass spectrometry-based metabolomic ap- from mussel tissue using laser ablation electrospray ion- proach for the characterization of goat milk compared with cow ization—tandem mass spectrometry,” Toxicon, vol. 92, milk,” Journal of Dairy Science, vol. 97, no. 10, pp. 6057–6066, pp. 75–80, 2014. 2014. [177] M. Guijarro-D´ıez, L. Volzal, M. L. Marina, and A. L. Crego, [191] L. L. Monti, C. A. Bustamante, S. Osorio et al., “Metabolic “Metabolomic fingerprinting of saffron by LC/MS: novel profiling of a range of peach fruit varieties reveals high authenticity markers,” Analytical and Bioanalytical Chem- metabolic diversity and commonalities and differences istry, vol. 407, no. 23, pp. 7197–7213, 2015. during ripening,” Food Chemistry, vol. 190, pp. 879–888, [178] L. Millan,´ M. C. Sampedro, A. Sanchez et al., “Liquid 2016. chromatography–quadrupole time of flight tandem mass [192] M. N. A. Khalil, M. I. Fekry, and M. A. Farag, “Metabolome spectrometry–based targeted metabolomic study for varietal based volatiles profiling in 13 date palm fruit varieties from discrimination of grapes according to plant sterols content,” Egypt via SPME GC-MS and chemometrics,” Food Chem- Journal of Chromatography A, vol. 1454, pp. 67–77, 2016. istry, vol. 217, pp. 171–181, 2017. [179] M. Arbulu, M. C. Sampedro, A. Gomez-Caballero, M. A. Goicolea, [193] I. Akhatou, R. Gonz alez-Dom´ınguez, and A. Fern andez- and R. J. Barrio, “Untargeted metabolomic analysis using liquid Recamales, “Investigation of the effect of genotype and chromatography quadrupole time-of-flight mass spectrometry for agronomic conditions on metabolomic profiles of selected non-volatile profiling of wines,” Analytica Chimica Acta, vol. 858, strawberry cultivars with different sensitivity to environ- pp. 32–41, 2015. mental stress,” Plant Physiology and Biochemistry, vol. 101, [180] J. Rubert, O. Lacina, M. Zachariasova, and J. Hajslova, pp. 14–22, 2016. “Saffron authentication based on liquid chromatography [194] A. Cuadros-Ivolstroza, S. Ru´ız-Lara, E. Gonz alez, A. Eckardt, high resolution tandem mass spectrometry and multivariate L. Willmitzer, and H. Pena-Cortes, “GC-MS metabolic profiling data analysis,” Food Chemistry, vol. 204, pp. 201–209, 2016. of Cabernet Sauvignon and Merlot cultivars during grapevine [181] A. Karlund,˚ U. Moor, G. McDougall, M. Lehtonen, berry development and network analysis reveals a stage- and R. O. Karjalainen, and K. Hanhineva, “Metabolic profiling cultivar-dependent connectivity of primary metabolites,” Metabolomics, vol. 12, no. 2, p. 39, 2016. discriminates between strawberry (Fragaria × ananassa [195] B. Khakimov, R. J. Mongi, K. M. Sørensen, B. K. Ndabikunze, Duch.) cultivars grown in Finland or Estonia,” Food Research B. E. Chove, and S. B. Engelsen, “A comprehensive and International, vol. 89, pp. 647–653, 2016. comparative GC-MS metabolomics study of non-volatiles in [182] Z. Jandric, D. Roberts, M. N. Rathor, A. Abrahim, M. Islam, Tanzanian grown mango, pineapple, jackfruit, baobab and and A. Cannavan, “Assessment of fruit juice authenticity tamarind fruits,” Food Chemistry, vol. 213, pp. 691–699, 2016. using UPLC-QToF MS: a metabolomics approach,” Food [196] J. Welzenbach, C. Neuhoff, C. Looft, K. Schellander, Chemistry, vol. 148, pp. 7–17, 2014. E. (olen, and C. Große-Brinkhaus, “Different statistical [183] Z. Jandric, M. Islam, D. K. Singh, and A. Cannavan, “Au- approaches to investigate porcine muscle metabolome thentication of Indian citrus fruit/fruit juices by untargeted profiles to highlight new biomarkers for pork quality as- and targeted metabolomics,” Food Control, vol. 72, sessment,” PLoS One, vol. 11, no. 2 article e0149758, 2016. pp. 181–188, 2017. [197] G. Min-Lee, D. Ho-Suh, E. Sung-Jung, and C. Hwan-Lee, [184] S. Jin, C. Song, S. Jia et al., “An integrated strategy for es- “Metabolomics provides quality characterization of com- tablishment of curcuminoid profile in turmeric using two mercial gochujang (fermented pepper paste),” Molecules, LC–MS/MS platforms,” Journal of Pharmaceutical and vol. 21, no. 7, p. 921, 2016. Biomedical Analysis, vol. 132, pp. 93–102, 2017. [198] D. E. Lee, G. R. Shin, S. Lee et al., “Metabolomics reveal that [185] K. Inoue, C. Tanada, T. Sakamoto et al., “Metabolomics amino acids are the main contributors to antioxidant activity approach of infant formula for the evaluation of contami- in wheat and rice gochujangs (Korean fermented red pepper nation and degradation using hydrophilic interaction liquid paste),” Food Research International, vol. 87, pp. 10–17, 2016. chromatography coupled with mass spectrometry,” Food [199] E. J. Gu, D. W. Kim, G. J. Jang et al., “Mass-based metab- Chemistry, vol. 181, pp. 318–324, 2015. olomic analysis of soybean sprouts during germination,” [186] M. Mattarozzi, M. Milioli, F. Bianchi et al., “Optimization of Food Chemistry, vol. 217, pp. 311–319, 2017. a rapid QuEChERS sample treatment method for HILIC- [200] C. Sales, M. I. Cervera, R. Gil, T. Portoles, E. Pitarch, and MS2 analysis of paralytic shellfish poisoning (PSP) toxins in J. Beltran, “Quality classification of Spanish olive oils by mussels,” Food Control, vol. 60, pp. 138–145, 2016. untargeted gas chromatography coupled to hybrid [187] A. K. Subbaraj, Y. H. Brad-Kim, K. Fraser, and M. M. Farouk, quadrupole-time of flight mass spectrometry with atmo- “A hydrophilic interaction liquid chromatography–mass spheric pressure chemical ionization and metabolomics- spectrometry (HILIC-MS) based metabolomics study on based statistical approach,” Food Chemistry, vol. 216, colour stability of ovine meat,” Meat Science, vol. 117, pp. 365–373, 2017. pp. 163–172, 2016. [201] D. K. Trivedi, K. A. Hollywood, N. J. W. Rattray et al., “Meat, [188] G. Aliakbarzadeh, H. Sereshti, and H. Parastar, “Pattern the metabolites: an integrated metabolite profiling and lip- recognition analysis of chromatographic fingerprints of idomics approach for the detection of the adulteration of Crocus sativus L. secondary metabolites towards source beef with pork,” Analyst, vol. 141, no. 7, pp. 2155–2164, 2016. identification and quality control,” Analytical and Bio- [202] V. D. Daygon, S. Prakash, M. Calingacion et al., “Un- analytical Chemistry, vol. 408, no. 12, pp. 3295–3307, 2016. derstanding the jasmine phenotype of rice through metab- [189] F. R. Pinu, S. de Carvalho-Silva, A. P. Trovatti Uetanabaro, and olite profiling and sensory evaluation,” Metabolomics, S. G. Villas-Boas, “Vinegar metabolomics: an explorative study vol. 12, no. 4, p. 63, 2016. 24 Journal of Analytical Methods in Chemistry

[203] L. Di Donna, D. Taverna, S. Indelicato, A. Napoli, G. Sindona, and F. Mazzotti, “Rapid assay of resveratrol in red wine by paper spray tandem mass spectrometry and isotope dilution,” Food Chemistry, vol. 229, pp. 354–357, 2017. [204] H. V. Pereira, V. S. Amador, M. M. Sena, R. Augusti, and E. Piccin, “Paper spray mass spectrometry and PLS-DA improved by variable selection for the forensic discrimination of beers,” Analytica Chimica Acta, vol. 940, pp. 104–112, 2016. [205] J. A. Reis Teodoro, H. V. Pereira, M. M. Sena, E. Piccin, J. J. Zacca, and R. Augusti, “Paper spray mass spectrometry and chemometric tools for a fast and reliable identification of counterfeit blended Scottish whiskies,” Food Chemistry, vol. 237, pp. 1058–1064, 2017. [206] A. K. Meher and Y. C. Chen, “Analysis of volatile compounds by open-air ionization mass spectrometry,” Analy- tica Chimica Acta, vol. 966, pp. 41–46, 2017. [207] M. Ha, J. H. Kwak, Y. Kim, and O. P. Zee, “Direct analysis for the distribution of toxic glycoalkaloids in potato tuber tissue using matrix-assisted laser desorption/ionization mass spectrometric imaging,” Food Chemistry, vol. 133, no. 4, pp. 1155–1162, 2012. [208] S. Taira, S. Shimma, I. Osaka et al., “Mass spectrometry imaging of the capsaicin localization in the Capsicum fruits,” International Journal of Biotechnology for Wellness In- dustries, vol. 1, pp. 61–66, 2012. [209] Y. Yoshimura, H. Evolmoto, T. Moriyama, Y. Kawamura, M. Setou, and N. Zaima, “Visualization of anthocyanin species in rabbiteye blueberry Vaccinium ashei by matrix- assisted laser desorption/ionization imaging mass spectrometry,” Analytical and Bioanalytical Chemistry, vol. 403, no. 7, pp. 885–1895, 2012. [210] A. Yasuda, Y. Tatsu, and Y. Shigeri, “Characterization of triacetyl-α-melanocyte-stimulating hormone in carp and goldfish,” General and Comparative Endocrinology, vol. 175, no. 2, pp. 270–276, 2012. [211] M.-Z. Huang, S.-C. Cheng, S.-S. Jhang et al., “Ambient molecular imaging of dry fungus surface by electrospray laser desorption ionization mass spectrometry,” In- ternational Journal of Mass Spectrometry, vol. 325–327, pp. 172–182, 2012. [212] S. S. Jhang, M.-Z. Huang, and J. Shiea, “Ambient molecular imaging of toxins within a sprouted potato slice by ELD- I/MS,” in Proceedings of the 60th ASMS Conference on Mass Spectrometry and Allied Topics, Vancouver, BC, Canada, May 2012. [213] R. Garrett, C. M. Rezende, and D. R. Ifa, “Revealing the spatial distribution of chlorogenic acids and sucrose across coffee bean endosperm by desorption electrospray ionization-mass spectrometry imaging,” LWT-Food Science and Technology, vol. 65, pp. 711–717, 2016. [214] P. M. Kumara, A. Srimany, G. Ravikanth, R. U. Shaanker, and T. Pradeep, “Ambient ionization mass spectrometry imaging of rohitukine, a chromone anti-cancer alkaloid, during seed development in Dysoxylum binectariferum Hook.f (Meliaceae),” Phytochemistry, vol. 116, pp. 104–110, 2015. [215] M. W. Nielen and T. A. van Beek, “Macroscopic and microscopic spatially-resolved analysis of food contaminants and constituents using laser-ablation electrospray ionization mass spectrometry imaging,” Analytical and Bioanalytical Chemistry, vol. 406, no. 27, pp. 6805–6815, 2014. Hindawi Journal of Analytical Methods in Chemistry Volume 2018, Article ID 1292954, 20 pages https://doi.org/10.1155/2018/1292954

Research Article High-Precision In Situ 87Sr/86Sr Analyses through Microsampling on Solid Samples: Applications to Earth and Life Sciences

Sara Di Salvo,1 Eleonora Braschi ,2 Martina Casalini,1 Sara Marchionni,1 Teresa Adani,1 Maurizio Ulivi,1 Andrea Orlando ,2 Simone Tommasini,1 Riccardo Avanzinelli,1,2 Paul P. A. Mazza,1 Sandro Conticelli ,1,2 and Lorella Francalanci1,2

1Dipartimento di Scienze della Terra, Universit`a degli Studi di Firenze, via Giorgio La Pira 4, 50121 Firenze, Italy 2C.N.R., Istituto Geoscienze e Georisorse, U.O. di Firenze, via Giorgio La Pira 4, 50121 Firenze, Italy

Correspondence should be addressed to Eleonora Braschi; [email protected]

Received 15 December 2017; Accepted 18 February 2018; Published 22 April 2018

Academic Editor: Veronica Termopoli

Copyright © 2018 Sara Di Salvo et al. )is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. An analytical protocol for high-precision, in situ microscale isotopic investigations is presented here, which combines the use of a high-performing mechanical microsampling device and high-precision TIMS measurements on micro-Sr samples, allowing for excellent results both in accuracy and precision. )e present paper is a detailed methodological description of the whole analytical procedure from sampling to elemental purification and Sr-isotope measurements. )e method offers the potential to attain isotope data at the microscale on a wide range of solid materials with the use of minimally invasive sampling. In addition, we present three significant case studies for geological and life sciences, as examples of the various applications of microscale 87Sr/86Sr isotope ratios, concerning (i) the pre-eruptive mechanisms triggering recent eruptions at Nisyros volcano (Greece), (ii) the dynamics involved with the initial magma ascent during Eyjafjallajokull¨ volcano’s (Iceland) 2010 eruption, which are usually related to the precursory signals of the eruption, and (iii) the environmental context of a MIS 3 cave bear, Ursus spelaeus. )e studied cases show the robustness of the methods, which can be also be applied in other areas, such as cultural heritage, archaeology, petrology, and forensic sciences.

1. Introduction magma residence times prior to eruptions (e.g., [1–10]). Since crystals record changes occurring in the environment In situ radiogenic isotope determinations with microscale in which they grow (e.g., [7, 11–15]), isotopic investigations resolution, especially of Sr, can represent a powerful tool in at grain and subgrain scales on rock-forming minerals different fields of geological and life sciences. In particular, provide information on mineral-whole rock equilibria that this technique is nowadays one of the most important constrain the magmatic processes occurring during magma methods for the investigation and interpretation of magmatic evolution (e.g., mixing, mingling, crystals recycling, crustal processes, as well as of environmentally-induced responses of contamination, or metasomatism). In addition, variation of terrestrial mammals; it has the potential to greatly enhance Sr-isotope composition from core to rim within the same our understanding of not only volcanic systems and the re- crystal can shed light on the complex pre-eruptive history lated magma genesis and evolution, but also of the physio- of active volcanoes. )erefore, combining micro-(small- logical mechanisms behind specific organic adaptations. scale) isotope data with textural and petrographic data In situ 87Sr/86Sr provides significant data on the (i) provides significant information on crystals residence source heterogeneities of magmas, (ii) crystallization his- time, magma production rates, and recharge dynamics tories within shallow level magmatic reservoirs, and (iii) (e.g., [5–8, 16–18]). 2 Journal of Analytical Methods in Chemistry

Sr-isotopic investigation has recently gained popularity Florence University is equipped with a modern MicroMill™ in other fields, such as archaeology, anthropology, biology, grinder device, an ultraclean laboratory (“Class 1000”) for cultural heritage, environmental and food studies, forensics, microsamples digestion and elemental separation and a ther- life and medical sciences, and palaeontology. (e.g., [19–38]). mal ionisation mass spectrometer ()ermoFinningan™ 87Sr/86Sr on tooth enamel and bone tissues gained particular Triton-Ti®) for isotopic measurements. attention over the last years in life studies, archaeology, Compared to classic, bulk sample analyses (generally palaeontology, and forensic sciences (e.g., [39–48]). In situ measured on 100–150 ng of Sr [55]), small-sample analysis analyses of 87Sr/86Sr using laser ablation and multicollector- (typically 5 to 10 ng of Sr) has the drawback of being more inductively coupled plasma mass spectrometry (LA-MC- exposed to contamination from Sr alien to the sample. In situ ICPMS) for tooth enamel, bones, and rocks were developed micro-Sr measurements therefore require continued testing in the early new millennium but have met variable success of laboratory blanks during the whole analytical procedure. (e.g., [49–53]). At the same time, microscale sampling through drilling and micro-Sr isotope analyses by thermal ionisation mass spectrometry (TIMS) was also developed on geological 2.1. In Situ Sampling materials (e.g., [6, 18, 54]). As compared with mechanical microdrilling plus TIMS procedures, LA-MC-ICPMS has 2.1.1. +e Microdrilling Device. Microsampling on minerals, the advantage of significantly reducing the time of analyses, glasses, and tooth enamel reported in this paper was per- but at the expense of precision of the measured isotope ratio; formed using a microdrilling device capable of high resolution this is generally due to smaller Sr signals and the need to milling (New Wave-Merchantek MicroMill™, https://www. correct for isobaric interferences (e.g., [16, 53]). esi.com/products/laser-processing/milling/micromill/). )e We present a detailed protocol for in situ sampling MicroMill (Figure 1(a)) combines a binocular microscope through microdrilling, Sr purification, and thermal ionisa- (with 6.7x–40x magnifying power) with remotely controlled tion mass spectrometer (TIMS) high-precision determina- submicron stage resolution and positional accuracy and a real- tions of small amounts of Sr (<10 ng) in both biological and time video observation (at 3x digital magnification). It in- geological materials at the same error levels. In situ 87Sr/86Sr cludes a low-eccentricity high-torque milling chuck, with analysis is presented in three case studies which deal with the variable speed (1,200–35,000 rpm), wherein a tungsten car- following: (i) plagioclase crystals from Nisyros volcano bide or diamond-tipped bit is fixed, and an automated high- (Greece), (ii) glassy matrix of single glassy clasts from the precision sliding stage on which the sample is loaded (Figure 1). 2010 eruption of the Eyjafjallajokull¨ volcano (Iceland), and )e open stage architecture can accommodate thin sections or (iii) bones and teeth from fossil remains of the Ursus spe- mounts and also larger solid samples such as bones, shells, laeus. In these three examples, the in situ Sr-isotope ap- ceramics, and plastics. )e plug where the bit is hosted moves proach permits to constrain petrological and volcanological with adjustable speed for both spin and vertical movement, processes and to effectively outline the life habits and habitat along the Z direction. )e stage moves along the X-Y direction exploitation of extinct living species. )e three presented with a precision of 1 μm and maximum shift of 5 cm, allowing cases aim to show the high potential of the methodology, a high spatial resolution to the micron scale. )is yields high validating its wide-scale applicability in many other areas, spatial resolution, to the micron scale, and allows small-size such as cultural heritage, archaeology, and forensic sciences. sampling (i.e., a few μg of powder). )e digital camera, placed next to the milling chunk (Figure 1(c)), shows a live image 2. Materials and Methods of the stage and of the bit position (Figure 2(a)). Using the microscope position mode (scope position), the optical image Technological improvements on mass spectrometer and of the sample can be monitored on the PC screen prior to and microsampling devices allow researchers to collect and after the sampling (Figure 2(b)). analyse small amounts of sample (few micrograms of )e device allows in situ microsampling on several types sample, containing nanograms of Sr) with no loss of pre- of solid materials such as rocks, minerals, glasses, plastics, cision in the isotopic determination. In situ analyses have bone tissues, ceramics, metals, and alloys. )e designed many important advantages over more typical ones on bulk software package also allows performing different milling samples. In geological applications, it allows to preserve the patterns such as holes arranged randomly, lines, or rasters textural information and thus to combine it with the isotopic (Figure 2(b)). Fine adjustment of milling velocity helps the and geochemical composition of specific portions of the microsampling of solid materials with different hardness. samples. In archaeology and palaeontology, this method has Microsampling on geological (e.g., rock, minerals, and the advantage of minimising the damage and/or destruction glasses) and archaeological materials (glasses, ceramics, teeth, of samples, thereby leaving significant amounts available for and bones) is performed using either thick polished sections further applications. (some 100 µm thick) or sample mounts (e.g., small chunks, )e procedure consists in three main stages: (i) in situ chips, fragments of bones, and teeth). In our case studies, we sampling through microdrilling, (ii) sample digestion and used thick, polished sections for mineral and volcanic glasses purification of the element of interest, in our case Sr, and (iii) and mounts for the teeth; both were fixed on the stage using measurement of the isotope ratios (i.e., 87Sr/86Sr) through either a double-sided adhesive tape or hot glue, to avoid thermal ionisation mass spectrometry (TIMS). Our Radiogenic sample displacement during milling. )ick, polished sections Isotope Laboratory at the Department of Earth Sciences of the are preferable for geological samples because they permit to Journal of Analytical Methods in Chemistry 3

(b)

(a) (c)

Figure 1: (a) )e New Wave-Merchantek Micromill device operating at the Department of Earth Sciences–University of Firenze; (b) image of a petrographic polish thick section ﬁxed on the sample stage under the tungsten carbide drilling bit that is locked into the milling chuck. A Milli-Q droplet constrained by the Paraﬁlm is placed on the section in order to collect the powder during the drill; and (c) sample slurry recovery from the drilled surface into the digestion beaker.

Raster 2: 27 line scans

Raster 1: 12 line scans Line of spot: 500 µm 28 drilling points

(a) (b)

Figure 2: (a) Live image of the sample stage and drill bit shown by the digital camera placed next to the milling chuck; (b) microscope view image showing a zoned plagioclase crystal with drilling pattern. Two rasters (in yellow) are set up to drill the plagioclase core, whereas a line of spot is set up for the rim microdrilling. characterise the petrographic features of the samples and thus a small punched square of warmed-up Paraﬁlm™ on the to perform the microsampling according to their textural sample surface (Figure 1(c)). )e water droplet retains the properties. powder produced by the milling, which can then be easily collected by pipetting; it has also the eﬀect of cooling the 2.1.2. Milling Procedure and Sample Collection. A droplet of microdrill bit while milling. Before each drilling session, the Milli-Q® water is placed with a micrometric pipette on the drill bit is ultrasonically cleaned with pure ethanol and then selected area prior to milling; this is performed by sticking rinsed with Milli-Q water. 4 Journal of Analytical Methods in Chemistry

Single conical hole Single line scan r r L

h h δ =15º δ =15º

200 µm EHT = 20.00 kV Signal A = SE1 Date :22 Jan 2016 V = 1/3pr2h V = L ∗ h ∗ r + (1/3pr2h) WD = 10.0 mm Mag = 55 X Time :10:02:11

(a) (b)

Figure 3: (a) SEM imaging of a tungsten carbide drill bit used for milling: the tip angle is 30° and geometrical shape of (b) a single conical hole and single line scan.

)e instrument software provides different drilling 2.2. Sample Dissolution and Sr Purification. )e purification patterns and depths: single or multiple independent holes, of the element of interest, in our case Sr, is crucial to obtain spot lines, grids, line scans, or rasters (Figure 2(b)). Milling high-precision isotopic measurements for at least two reasons. spot lines (or grids) are more accurate, but more time- First, it avoids isobaric interferences on the masses that will be consuming; they were used for the geological material analysed by mass spectrometry; in the specific case of Sr (crystals and volcanic ashes) which requires more precise isotope measurements, even a small amount of 87Rb will add spatial resolution between the different zones of the same to 87Sr, yielding an overestimate of the 87Sr/86Sr ratio. Sec- crystal or between the thin films of glass. Milling failure, such ondly, the presence of other elements of the matrix will as crystal breaking, was prevented by setting a slow scan compete with Sr during the thermal ionisation process, re- speed and splitting the milling into two or more steps. Line ducing the Sr signal and thus yielding less accurate mea- scans, which are performed faster but less accurately, were surements. )e possibility of collecting and processing the used for drilling the teeth. )e number of points, lines, or sample for the purification of the element of interest is one of rasters to be milled (which accounts for the amount of Sr to the major advantages of the method presented here over other be collected) need to obtain a sufficient quantity of Sr for the methodologies, which do not achieve the same degree of TIMS measurements and can be calculated based on (i) the accuracy and precision. )e LA-MC-ICPMS methods allow Sr content of the sample (independently determined by faster data acquisition and higher sample throughput than LA-ICPMS), (ii) the geometry of the drill bit, and (iii) the mechanical microdrilling plus TIMS procedures, thanks to drilling pattern and the depth. Tips of different size and shape the possibility of introducing the samples directly into the can be used for the drilling; therefore, the volume of material mass spectrometer without chemical separation. On the other actually removed from different depths of a single hole or at hand, LA-MC-ICPMS measurements require careful moni- different depths and lengths of a single line needs to be toring and corrections to minimize isobaric interferences carefully calculated. )e tungsten carbide mill bits supplied in order to achieve suitable analytical accuracy and precision with the microdrill device (Komet–Brassler), have conical (e.g., [49, 53, 54, 56]). shape with an angle of 30° (Figure 3). )e volume removed Powder digestion and Sr purification were carried out in during each drilling is equivalent to that of the conical tip and our ultraclean laboratory (“Class 1000”) aiming at the fol- dependent on the geometry of the drilling pattern, as well as lowing: (i) optimising the separation of Sr and Rb to avoid on the specific depth (Figure 3). )e minimum amount of interference of 87Rb with 87Sr, (ii) purifying the Sr collection sample that needs to be drilled depends also on the total form all the matrix analytes, (iii) maximising the yield of the procedural blank, which should be at least two orders of columns during the chromatographic purification, and (iv) magnitude lower than the total amount of Sr collected from preserving low procedural blanks. Sample digestion was the sample. performed by sequential HF-HNO3-HCl as described in After milling was completed, the sample slurry was col- [55]. Chromatographic Sr purification was performed using lected with a micropipette in a PFA beaker and then trans- Eichrom® Sr-Spec™ resins (100–150 µm) in quartz micro- ferred in the clean lab for sample digestion and elemental columns (0.14 ml volume; Figure 4). Matrix elements were purification. )e blanks of the milling procedure were de- flushed out through elution with 14 column volumes of 3 N termined by keeping the drill bit tip into a Milli-Q water HNO3. Sr was then collected in Milli-Q (13 column vol- droplet on the sample surface (accurately cleaned before use) umes). )e collected Sr fractions were further treated with ° for as long as the average sampling time; the droplet was then concentrated HNO3 and H2O2 (fluxing at 150 C on a hot- processed as an ordinary sample. )e amount of Sr in the plate) to remove any organic residue. After this final step, blank was then determined through isotope dilution, by samples were diluted in HNO3 (10 vol.%) and were finally adding a single-spike solution (enriched in 84Sr). ready for loading on filaments for mass measurements. )e Journal of Analytical Methods in Chemistry 5

(a) (b)

Laminar ﬂow hood

Microcolumn holder

2 cm

Figure 4: (a) Image of the laminar flow hood used for chemical processing and Sr separation of microdrilled samples; (b) image of quartz microcloumns filled with approximately 140 µl of specific chromatographic resin (Eichrom Sr-Spec, 100–150 µm) with high Sr-recovery efficiency, for Sr element extraction.

Table 1: Blank contamination level. Laboratory blanks Isotope ratios on blanks Sr pg 1 SD n 87Sr/86Sr 1 SD n Standard procedure on large size samples 127 60 32 0.707497 0.000060 5 Procedure on small size samples Total procedure (drilling, digestion, 38 19 16 and elemental selection) Chemical digestion and separation 17 6 12 1 SD, standard deviation (external precision); n, numbers of blank measurements; the average Sr-isotope value obtained from 5 unspiked blanks is reported to fully characterize the potential contamination component. whole analytical procedure was performed with acids of multidynamic, e.g., [55, 59]). Both are very important to ultra-pure quality. maximise the Sr signal during the measurements, to balance In order to thoroughly assess the contamination levels, we the analysis time and the analytical errors that are a function measured two types of blanks, one considering only the of sample size. amount of Sr deriving from the chemical digestion and Sr )e measurement protocol was tested by replicate analyses separation, the other accounting for the whole procedure, of an international certified standard (NIST-SRM987), prop- including the drilling process, as described in the previous erly diluted to attain sample sizes (5 to 10 ng Sr) comparable section. )e results were 17 ± 6 (1 SD, n � 12) and 38 ± 19 pg to those of the microsamples. )en, we tested the whole (1 SD, n � 16), respectively, over a 14-month period (Table 1), procedure, from in situ sampling (<10 ng of Sr) to isotope thus allowing sampling as low as 4 ng of Sr for isotope analysis. measurement, on the international glass reference sample BHVO-2G. BHVO-2G reference sample is a synthetic basal- 87 86 tic glass (provided by USGS) obtained by melting the BHVO- 2.3. Sr/ Sr Measurements on TIMS. Sr isotope ratios were 2 powder collected from a Hawaiian lava flow. )e glassy determined using a multicollector, thermal ionization mass slices are supplied in epoxy resin mounts (https://crustal. spectrometer (TIMS: )ermoFinningan Triton-Ti ) (Figure 5), ™ usgs.gov/geochemical_reference_standards/microanalytical_ equipped with nine moveable collectors, which allow to 84 86 RM.html). simultaneously detect all the natural masses of Sr ( Sr, Sr, 87Sr, and 88Sr). )e mass of 85Rb was also measured to monitor possible 87Rb interference, but it was always lower than the 2.3.1. Sample Loading onto Filament. )e Sr fraction col- detection limit of the instrument, confirming the quality of the lected from the columns was dissolved into 1 µl HNO3 separation procedure described above. A detailed description 10 vol.% and loaded on single Re filaments under a hori- of instrumental characteristics and performances are given zontal laminar flow hood. Due to the small amount of Sr in [55], along with standardised routine, measuring con- available, it is important to confine the sample on the ditions, and setting for normal-sized samples (100–150 ng smallest possible area on the filament, so that the whole of Sr). Instrumental mass bias (e.g., [57–59]) was corrected loaded sample can be ionised at the same time from a single to the natural value of 86Sr/88Sr � 0.1194 using an exponential spot. To attain this, a thin layer of Parafilm™ was melted at law (e.g., [55, 59]). both sides of the filament surface, to prevent any spreading )e most critical aspects of measuring small-size samples of the solution, leaving a small gap (1 mm) for the droplet at are related to (i) the procedure of sample loading onto the the center on the filament (Figure 5(b)). )e loading was filaments and (ii) the measurement mode (i.e., static versus performed by sandwiching the sample between two 0.5 µl 6 Journal of Analytical Methods in Chemistry

(a)

Sample droplet

Rhenium ﬁlament with dried sample Rhenium ﬁlament Filament-holder turret

(b) (c) (d)

Figure 5: (a) )e )ermoFinningan Triton-Ti multicollector, thermal ionization mass spectrometer (TIMS) at the University of Firenze; (b) single-rhenium filament with a sample droplet on top; (c) detail of a single filament, holding a dried sample loaded on a sample holder turret that can host 21 filament positions; (d) the turret is ready for the installation into the thermal ionization mass spectrometer. drops of TaCl5 activator solution [55] and 0.5 µl of H3PO4 )e Triton-Ti is equipped with a virtual amplifier, which en- solution (6 vol.%), respectively. TaCl5 activator was added ables a variable connection between amplifiers and Faraday for enhancing Sr ionisation efficiency and H3PO4 for sta- cups and allows a complete switching between amplifiers and bilising Sr isotope fractionation during measurement. All the cups during a single measurement. However, the virtual am- solutions (i.e., activator, sample, and H3PO3) were slowly dried plifier is not able to correct for the different Faraday cup ef- by passing a current on the filament, which was increased at ficiency and its variation with time. the end of the procedure until the filament starts glowing. In contrast, the dynamic (or multidynamic) mode is )e loaded filaments were then placed on the filament-holder a peak-jumping procedure where a number of different cup turret and then inserted into the mass spectrometer (Figures configurations are employed for determining a single iso- 5(c) and 5(d)). topic ratio (Table 2). )is means that each isotope beam is measured sequentially in different Faraday cups, so that two 87 86 Sr/ Srdouble values (Table 2) can be calculated without cup 2.3.2. Measurement Procedure Reproducibility and Accuracy. efficiency biases and drifts of the electronics (Table 2). )e 87 86 )e measurement routine was established to obtain the best two Sr/ Srdouble values are then geometrically averaged to 87 86 87 86 internal and external precisions, and the accuracy, on Sr/ Sr, obtain a single Sr/ Srtriple value. was achieved by experimentally comparing runs performed )e best configuration for static mode measurements in static versus dynamic conditions at a variable number of was found by measuring 300 cycles with an integration time cycles and integration times. A detailed description of the of 8 s, which corresponds to a total measuring time of about static and dynamic methods is provided in [55, 59]. In brief, 35 minutes for each sample. For dynamic mode measure- the static mode consists of simultaneous measurements of all ments, we performed 120 cycles (each including 3 magnetic isotopes in a single “jump,” so that the magnetic field remains jumps), with 8 s of integration time and an idle time of 3 s static and the masses always hit the same detectors (Table 2). between the different jumps, for a total of 70 minutes for Static measurements have the advantage of considerably re- each sample. ducing the acquisition time in comparison with the dynamic In both static and dynamic methods, the filament was mode, which becomes important when little amount of Sr is slowly warmed up, for a total of about 45 minutes, to stabilise available, as dealing with small samples. )e main limitation of the ion emission until the suitable intensity is achieved. this method is related to the uncertainty on the Faraday cup During the heating, the beam was accurately optimised by efficiency and on the drift of the electronics (i.e., the amplifiers). peak-centering and focusing. )e optimal beam intensity for Journal of Analytical Methods in Chemistry 7

Table 2: Cup configuration schemes of static (a) and dynamic (b) mode measurements. Cup L4 L3 L2 L1 C (Far) H1 H2 H3 H4 (a) Static collection mode 84Sr 85Rb 86Sr 87Sr 88Sr (b) Dynamic collection mode Jump 1 85Rb 85Rb 86Sr 87Sr 88Sr Jump 2 (main) 84Sr 86Sr 86Sr 87Sr 88Sr Jump 3 87Sr 88Sr Sr double Combining the measurements from two different magnetic field�� position (jump 1-2 and jump 2-3 above), it is possible to evaluate two 87 86 [87 /86 ] � [87 /86 ] ·[87 /88 ] ·[88 /86 ] [87 /86 ] � independent��Sr/ Srdouble defined as follows: Sr Sr 1-2 SrH1 SrC 1 SrC SrH1 2 Sr Sr N and Sr Sr 2-3 87 86 87 88 88 86 [ SrC/ SrL1]2 ·[ SrL1/ SrC]3 ·[ Sr/ Sr]N. )e numbers outside the parentheses are relative to the three different magnetic field 88 86 positions, the subscript of each isotope refers to the cup on which it is measured, and [ Sr/ Sr]N is the natural ratio (i.e., 8.375209) the measurement varied from static to dynamic mode, with significant differences in accuracy have been reported [53]. higher intensity allowed by the shorter static (3–3.5 V on On the other hand, the internal and external reproducibility 88Sr) mode with respect to dynamic (1.5–2 V on 88Sr) mode, worsen significantly (e.g., by a factor 5 to 10 in [16]) in which instead requires maintaining a stable beam, owing to materials with low Sr and high Rb/Sr. the longer duration of the measurement. In summary, the method presented here generally )e results are shown in Figure 6 and Table 3. Static and provides more accurate and precise results than LA-MC- dynamic mode measurements on NIST-SRM987 reference ICPMS, independently on the nature of the analysed ma- samples (10 ng of Sr measured) yielded 87Sr/86Sr average terial, despite being more time-consuming. It is therefore values of 0.710247 ± 0.000026 (2 SD, n � 30) and 0.710251 ± suitable for a wider range of applications. 0.000018 (2 SD, n � 51), respectively, with internal precisions of 13 ppm (2 SE) and 16 ppm (2 SE), respectively. 3. Applications Both values are within the recommended reference value 87 86 for NIST-SRM987 ( Sr/ Sr � 0.710248 ± 0.000011; Figures 6 In this section, we report three case studies, two of which were (a) and 6(b) and Table 3 [59]). Static measurement reduced previously published [8, 10], as examples of possible appli- the experimental time but showed a worse external re- cations of the presented methodology in different fields of producibility than that obtained in dynamic mode (Figures science. Indeed, in the last decade, the in situ isotope 6(a) and 6(b)), yet maintaining similar internal precision. microsampling approach has been used and applied in many )erefore, the dynamic mode was chosen for the experimental pilot studies in a wide range of research fields, including, work both on the international glass standard BHVO-2G and among others, palaeoenvironmental and palaeoecologic re- on the unknown samples. Further measurements of SRM987 constructions (i.e., [53, 63]) and climate changes (i.e., [64]). in the dynamic mode were performed after the initial testing, In the three presented case studies, the samples were along with the studied samples, yielding consistent results thoroughly characterised both texturally (optical micro- 87 86 ( Sr/ Sr � 0.710252 ± 0.000018, 2 SD, n � 47; Figure 6(c)). scope and SEM) and chemically (electron micro probe Results on the BHVO-2G are reported in Table 3. )e analyses) before drilling. )e strontium element concen- results were also compared to standard measurements (150 ng trations (in ppm) in all the samples was then determined of Sr) on the BHVO-2 powder reference sample. BHVO-2G through LA-ICPMS. versus BHVO-2 results found for micro- and normal-size samples, respectively, are well within the internal analytical error (Table 3) and in agreement with the reference values 3.1. Micro-Sr Isotope in Minerals. Rock-forming minerals in reported in [60, 61] for bulk powder (i.e., standard BHVO-2). igneous rocks display variable chemical composition depending )e significantly larger standard deviation of the micro- on several processes and parameters such as (i) the physi- drilled BHVO-2G measurements, with respect to both micro- cochemical conditions of the magmas, (ii) open system pro- Sr SRM987 and BHVO-2 powder data, is likely partly related cesses (e.g., magma mixing and mingling), and (iii) recycling to small isotopic heterogeneities of the glass standard. )e of cumulated crystals triggered by new arrivals of magma few available micro-Sr data on the same sample provide within crustal reservoirs. Radiogenic isotope ratios in minerals, similar averages and reproducibility of our data [54, 62] or portions of them, can be used as a petrogenetic “DNA” to (Table 3). record the history of the magma reservoir (crystal isotope Comparing our results with LA-MC-ICPMS data is stratigraphy, e.g., [7]) and their evolution within the crust (e.g., more difficult; in fact, the latter vary largely depending on [1, 5, 7, 8, 65–67]). Combining in situ Sr-isotope fingerprints the material used for the analyses. External reproducibility with other approaches, such as textural evidences and crystal obtained with LA-MC-ICPMS on material with high Sr size distribution, offers the opportunity to understand the contents and low Rb/Sr (e.g., apatite [49, 53], marine shells, processes and timescales through which magmas are stored, and synthetic plagioclase [16]) is comparable or slightly differentiated, and delivered prior to eruption (e.g., [5–7, worse than that attained with our method; yet, small but 13, 14, 16, 18, 54, 68, 69]). 8 Journal of Analytical Methods in Chemistry

0.71030

0.71028 20 irlwall [59] 0.710248 ± 11

0.71026 Sr

86 15 Sr/

87 0.71024

10 0.71022 irlwall [59] Runs # 0.710248 ± 11 Number of measurements 0.71020 5 10 20 30 40 50

Dynamic collection mode Static collection mode 87Sr/86Sr = 0.710251 87Sr/86Sr = 0.710248 2sd = 18 (n = 50) 2sd = 25 (n = 29) 0.710230 0.710270 2sd = 10 2sd = 10 m m Δ 87Sr/86Sr = 5 10–6

(a) (b) 0.71030 Average value: 87Sr/86Sr = 0.710252 ± 16 (n = 47)

0.71028

0.71026

Sr/ 87 0.71024

0.71022

0.71020

Time (reproducibility during 2014–2016 working period)

(c)

Figure 6: (a) Reproducibility and accuracy for repeated measurements of 87Sr/86Sr on 12 ng load size of NSIT-SRM987 standard material for static versus dynamic collection mode over a period of 10 months. Each single measurement is plotted with the relative error bars. )e grey- shaded field shows the [59] recommended value range. (b) Distribution of the 87Sr/86Sr values measured all over the period of analysis with the static and dynamic collection mode: the static collection mode (black filled columns) shows a worst external reproducibility than that obtained in the dynamic mode (open columns); indeed the dynamic mode gives measurements that reasonably fit a Gaussian distribution pattern with the more representative 87Sr/86Sr values centered within the [59] recommended interval (grey shade fields); (c) reproducibility and accuracy trend of 87Sr/86Sr 12 ng load size of NSIT-SRM987 measured in the dynamic collection mode throughout the setup period (from 2014 to 2016).

)e case study presented here is related to the active history, which is likely interpreted as the triggering Nisyros volcano, the easternmost volcanic island of the mechanism for its eruptions (e.g., [74, 77, 78]). Sr-isotope South Aegean Active Volcanic Arc (Figure 7) [5, 70–76]. determinations at the subcrystal scale, along with detailed Nisyros volcanic products are typically porphyritic rocks, petrographic microscopic textural evidence, provided with clear petrographic evidence of recurrent mixing and significant data for better defining the interaction of dif- mingling of different magmas during the whole volcano’s ferent magmas, concerning pre-eruptive mechanisms. )e Journal of Analytical Methods in Chemistry 9

Table 3: Accuracy and reproducibility on reference standard material (NIST-SRM987 and BHVO-2). Standard Sr content 87Sr/86Sr 2 SD n Reference Within run measurement on international standard NIST-SRM987 Large size loading (ng) SRM987 150 0.710253 0.000016 59 Long-term reproducibility (from 2013) SRM987 150 0.710248 0.000011 427 )irlwall [59] Small size loading SRM987 10 0.710252 0.000016 47 )is study SRM987 12 0.710259 0.000018 92 Charlier et al. [54] Drilling procedure on BHVO-2 glass BHVO-2 glass <10 0.703490 0.000092 9 )is study BHVO-2 glass 10 0.703492 0.000094 3 Charlier et al. [54] 87Sr/86Sr 2 SE BHVO-2 powder 150 0.703469 0.000004 1 )is study 87Sr/86Sr 2 SD BHVO-2 powder — 0.703479 0.000020 12 Weis et al. [60] 2 SD, two standard deviation (external precision); 2 SE, two standard error of the mean; n, numbers of measurements; literature data are reported in italics below our mean values for references. study focused, in particular, on postcaldera, rhyodacitic consequent drop in magma viscosity, thus favouring dome dome magmas of the final Nisyros activity emplaced after extrusion rather than explosive activity [8, 78]. the caldera-forming rhyolitic explosive eruption of upper pumice. )ese rhyodacitic lavas contain magmatic enclaves (Figures 7(a) and 7(c)), with basaltic andesite to andesite 3.2. Micro-Sr Isotope in Natural Glasses. Glasses are found in compositions interpreted, based on their textural features, nature generated by rapid quenching of molten material. as quenched portions of mafic magmas included in the )ey represent a volumetrically small component of crustal cooler, more evolved rhyodacitic host melt (Figures 7(b) rocks and can have different genesis (i.e., volcanic, lightning and 7(c)) (e.g., [8, 78]). strikes, meteorite impact, and anthropogenic). In this light, In situ Sr-isotope ratios were determined on plagioclase the radiogenic isotopic compositions (i.e., Sr, Nd, and Pb) phenocrysts (Figures 7(c)–7(e)), from both domes and en- can provide fundamental information to discriminate claves, which preserve evidence of the complex history of among the processes involved in their formation. Glasses interaction between the mafic (i.e., enclaves) and felsic (i.e., constitute the main component of ash and pyroclastic de- rhyodacitic domes) magmas in their growing zones [8]. )e posits, and their composition and Sr-isotope signature can 87Sr/86Sr values determined on micromilled samples from the provide important information in defining the triggering different growth zones of the plagioclase phenocrysts show mechanisms of explosive eruptions (e.g., [10, 69, 79–84]). clear Sr-isotope disequilibria between (i) cores and rims of Glasses may also be found in ceramics, as well as in other single crystals (Figures 7(c)–7(e)) and (ii) the crystals and the artefacts. Sr-isotope data are therefore also important to host magmas (Figure 7(e)). )is suggests that some of the define the possible source of raw materials for pottery, which phenocrysts that had formed in the rhyolitic magmas were is particularly relevant for cultural heritage, or to track trade later enclosed (as xenocrysts) within the more mafic one (i.e., routes in archaeology (e.g., [85–88]). Microscopic scale Sr-isotopic as enclaves). Whereas the rim of the phenocrysts is isotopically measurements on glasses would help in minimising the intermediate between the rhyolitic and mafic magmas (Figure amount of sample that needs to be milled, which is of crucial 7(f)), their cores show higher 87Sr/86Sr values, quite close to importance both dealing with small-sized volcanic ashes and that of the previously erupted upper pumice magma. )is ejecta and with human artefact of archaeological interest. clearly indicates that the phenocrysts originated in a different, )e present study is a case study of submicroscopic scale older system. In this light, the large plagioclase phenocrysts Sr-isotopic determination on ashes of the 2010 Eyjafjallajo¨kull found inside the dome lavas and enclaves can be interpreted as (Iceland) volcano’s explosive eruption, which caused enor- recycled from previously cumulated crystals (called “antecrysts” mous disruption to air travel across the northern hemi- by Davidson et al. [7]). sphere (Figure 8(a)) (e.g., [89–94]). )is eruption represents )ese results have also demonstrated that the dome a unique opportunity to test the potential of microscale Sr- lavas are multicomponent magmas formed by progressive isotope determinations on ash glasses from tephra deposits mingling/mixing processes between (i) a rhyolitic, and more that were well preserved within the ice/snow pack [10]. )e Sr-radiogenic melt derived from the original upper pumice population of tephra is composed by four different types of magmas, and (ii) the enclave-forming mafic, and less Sr- ash fragments: (i) fluidal, (ii) coarsely vesicular, (iii) radiogenic, melts refilling the felsic magma chamber. spongy fine vesicular, and (iv) blocky (Figure 8(b); [93]), )e constraints involved in interpreting in situ isotope data generated by different fragmentation processes during the have further implications for the timing and style of eruption. eruption. A detailed microanalytical geochemical and in )e inferred delay between the mafic input (i.e., enclaves) and situ Sr-isotope study performed on glassy groundmass of the relative dome eruption allows time for reheating and single ash clast showed unusually high 87Sr/86Sr values (up 10 Journal of Analytical Methods in Chemistry

100km Intracaldera Caldera rim N lava domes

Athens

Nisyros

(a)

Enclaves

Encalves Lava dome Lava dome

Plagioclase crystals

(b) (c) Microdrilled plagioclase within maﬁc enclaves

87Sr/86Sr

0.70397 0.7045 Upper pumice 0.70451 pyroclastites 0.7044 Core

0.7043

500µm 0.7042 Rhyodacitic (d) postcaldera lava domes

0.7041 (PCD) Rim 0.7040

0.7039 Basaltic andesite 0.70443 enclaves in the 0.7038 PCD 0.70398 Sr-isotope variation of the microdrilled cores and rims in plagioclase crystals 200 µm within domes and enclaves

(e) (f)

Figure 7: Results of micro-Sr isotope studies on plagioclase crystals from the last magmatic activity of Nisyros volcano (Greece). (a) Landscape view of the Nisyros caldera and its lava domes outpoured during the final magmatic activity of the volcano; (b) image of a lava dome outcrop rich in magmatic enclaves. Notably, the enclaves occurs as well-defined body with rounded and smooth surfaces; (c) specific image of a lava dome and enclave, both rich in large plagioclase crystals; (d and e) back-scattered electron microscope images of two representative plagioclase crystals selected for micro-Sr investigation with microdrill. )e shaded areas represent the crystal zones drilled for Sr-isotope analyses, both on cores and rims; (f) 87Sr/86Sr variation of the drilled cores and rims compared to the range of 87Sr/86Sr of host whole rock (domes and enclaves) and upper pumice whole rock. Journal of Analytical Methods in Chemistry 11

Ash clast types FL

SFV 500µm 200µm

CV BL 200µm 200µm

(a) (b) In situ Sr isotopes on single-ash glassy clasts of diferent types 0.707 D2

0.706

Sr 86

100µm 0.705 Sr/ 87 C3 (c) b1 + C1 0.704 C6 Range of Sr isotopes for the Iceland rocks

0.703 Drilled portions 59 60 61 62 63 64 of the clast SiO2 (wt.%.)

Clasts from Clasts from the opening-phase other phases of the eruption BL CV + BL 100µm BL CV CV + SFV + Microdrilled ash clast BL SFV SFV

(d) (e)

Figure 8: Results of micro-Sr isotopes in volcanic ash glasses of the 2010 explosive eruption of Eyjafjallajökullvolcano (Iceland). (a) Spectacular image of the ash cloud erupted by the volcano during the early stage of the 2010 activity (https://hiticeland.com/places_ and_photos_from_iceland/eyjafjallajokull causing); (b) back-scattered electron microscope images of different clast types sampled in the basal level of the fallout deposit related to the open phase of the 2010 eruption. FL � fluidal; SFV � spongy finely vesicular; CV � coarse vesicular; and BL � blocky. Scale bars are in micron; comparison between back-scattered electron microscope images of clasts before (c) and 87 86 after (d) in situ microdrilling. Scale bars are in micron; (e) Sr/ Sr versus SiO2 (wt.%) diagram of matrix glasses sampled from single, glassy ash clasts. In situ microsampling for Sr-isotope determinations on glass and plagioclase have been obtained by microdrilling technique. Error bars are inside the symbols. to 0.70668) for Icelandic volcanism (0.7026–0.7037 [94]) originated from the magma quenched from the contact (Figure 8(c)); these high isotopic values were also associated to with the ice cap filling the summit caldera of the volcano atypical elemental compositions compared to most of the [10]. )ese anomalous findings in the Icelandic mag- juvenile ash fragments of the eruption. )e anomalous, matic environment can be explained supposing that high Sr-radiogenic clasts belong to the blocky type (Figure 8 during its rise and before intruding into the ice cover, the (b)) and are concentrated in the first, thin ash level emplaced erupting magma selectively assimilated hydrothermal min- during the initial phase of eruptive activity. )ese clasts erals (i.e., zeolites, silica phases, and anhydrite) with seawater- 12 Journal of Analytical Methods in Chemistry related high-Sr isotopic ratios, hosted in altered volcanic/e- Ursus spelaeus was an endemic, widespread European piclastic rocks. )is selective assimilation took place at the tip Late Pleistocene species. In contrast to many other taxa, it edge of the first rising magma body, resulting in a high degree has fairly rich fossil records, especially thanks to its recurrent of contamination restricted to the rather small amount of melt use of caves or shelters for winter hibernation [114–116]. directly in contact with the hydrothermal veins. Indeed, Caves were frequently used by bears for many generations, evidence for this process is recorded only by the very first and numerous individuals eventually died in them, so that erupted juveniles (i.e., the blocky clasts). )e results obtained significant quantities of their remains accumulated over through submicroscopic-scale micromilling and relative considerable periods of time. Sr-isotope determination revealed the dynamics of the pro- Many studies used stable isotopes to determine the di- cesses involved in the initial stages of magma ascent to the etary habits of living and extinct bears (e.g., [117–121]), but surface; this provides significant insights into the in- until now only a small number of papers have considered terpretation of the precursory signals of the eruption (mostly employing Sr isotopes to possibly elucidate the factors consisting of ground deformation or increased seismicity) influencing habitat use and gain insights into the foraging [10]. )ese transient processes, which interested only a small, behaviour of cave bears [122]. We report here the first well-confined part of the magma, cannot be detected using 87Sr/86Sr data obtained through in situ microsampling on traditional, Sr-isotope determination on whole-rock samples teeth and bones of Ursus spelaeus found in Grotta all’Onda but can be revealed only analysing single-glassy clasts cave. )e study was aimed at defining the lifestyle and separately. feeding behaviour of one of the most prominent European Late Pleistocene mammals [123–125]. Grotta all’Onda cave is located 708 m above sea level (a.s.l.) 3.3. Micro-Sr Isotope in Teethand Bones. Due to their similar in the Apuan Alps nearby the village of Camaiore (Tuscany, chemical properties, Sr can substitute for Ca in the bio- Italy) (Figure 9), in a sub-Mediterranean habitat [126]. apatite [Ca5(PO4,CO3)3(OH,F)] of mammalian bones and )e cave opens at the base of the Tuscan Nappe, at the contact teeth, reaching contents of few hundreds of ppm that allows between the “Calcare a Rhaetavicula Contorta Formation” the isotope analysis by microdrilling. )e Sr isotope (i.e., Upper Triassic dolomitic-limestone of the Tuscan Nappe) composition of human and animal hard tissues is a func- and the “Argilliti Varicolori Formation” (i.e., Lower Cretaceous tion of their dietary habits (e.g., [22]) and depends on the shales) (Figure 9). )e Rhaetavicula Contorta Formation is isotopic composition of the food and water ingested during a polygenic breccia, mainly including metamorphic clasts, life, which in turn are related to the geological substrate known as “Brecce di Grotta all’Onda” (Figure 9) (http:// [95–97]. www502.regione.toscana.it/geoscopio/geologia.html). )e Sr isotopes have been successfully used, along with other fossil remains of Ursus spelaeus were recovered during stable and radiogenic isotope systematics (i.e., δ13C and δ18O a 1999 excavation. Radiocarbon dating of bone yielded ages and Pb isotopes), not only to track the source regions of ranging from 38.22 to 38.28 ky (BP) [127]. Six different migrants and migration pathways, as well as the hunting and specimens were selected for Sr isotope analysis; these in- trading areas of human populations, but also to study and clude three lower molars (SCT4, SCT5, and SCT6) from define the dietary habits of humans and animals (e.g., layers 7J4, 7J3, and 7J5, respectively and three metapodial [23, 44, 95–106]). bones, SCT1 (third metacarpal), SCT 2 (fourth metatarsal), A pioneering study [107] demonstrated that migrant and SCT3 (fourth metatarsal), from layers 7J4, 7J5, and 7J3, individuals who moved between different geologic regions respectively (Figures 9(c) and 10(a)–10(b)–10(c)–10(d)). might be traced by comparing 87Sr/86Sr in adult tooth Two whole soil (i.e., cave earth) samples, SCT7 and SCT8, enamel, formed between four and twelve years of age, and in were also collected from the representative layers 7J4 and the bones, which remodel throughout life and therefore 7J5 (Figure 9(c)). representative of adulthood [30]. Unlike bones and dentine, All the three teeth had well-preserved dentin and very dental enamel formed during childhood [108] remains thin enamel layer (Figure 10). In contrast, the three bones unaltered throughout the years. Different 87Sr/86Sr in the were rather differently preserved and had different porosity. teeth and bones of an individual may thus reflect the fact that In particular, SCT3 was the most heavily mineralised and it moved around the landscape passing through different best preserved, whereas SCT1 was densely vacuolated and isotopic environments during its youth and maturity preserved higher amounts of organic components. [109, 110]. Teeth enamel is generally preferred to dentine Major element analyses of these specimens revealed that and bones in the analysis of Sr concentrations and isotope the enamel bioapatite was more mineralised than that of ratios because it is virtually unaffected by postmortem dentin and bones. Moreover, the dentin bioapatite was diagenesis (e.g., [111–113]). found enriched in Sr and in other trace elements [128]. )e micro-scale Sr-isotope measurements of samples For the purpose of this study, dentin and enamel of the obtained using the submicroscopic-scale micromill tech- teeth and cortical sections of the bones from Grotta all’Onda nique is perfectly able to discriminate between enamel and were micromilled and analysed for micro-Sr isotope de- dentine in single-tooth samples. In addition, this technique termination. )e two soil samples were also processed for increases the accuracy of sampling and also reduces the Sr-isotope determination using the traditional, large sample amount of specimen to be destroyed for high-precision Sr- method [37, 55]. )e 87Sr/86Sr data are reported in Table 4. isotope analysis. Tooth enamel shows higher 87Sr/86Sr and lower Sr/Ca than Journal of Analytical Methods in Chemistry 13

Grotta all’Onda

SCT1 and 4 SCT3 and 5 SCT2 and 6

Grotta all’Onda entrance

(a) (b) (c) RSA Monte Matanna MAS Camaiore Grotta all’Onda

LIM

Alp Apuan

Camaiore

Tuscan Nappe Metamorphic units LIM RET PSM RSA ctc MCP MAS

(d)

Figure 9: )e Ursus spelaeus finding environment. (a) Image of the Grotta all’Onda cave from the above, showing the well-defined cut of the Mt. Matanna flank where the cave is located; (b) image of the entrance of the Grotta all’Onda cave; (c) image of the soil stratigraphy from which the Ursus spelaeus fossil remains have been sampled during the 1999 field campaign; (d) geological sketch map of the area around the Grotta all’Onda cave (http://www502.regione.toscana.it/geoscopio/geologia.html#). Formation of the Tuscan Nappe unit: LIM: “Calcare Secifero di Limano” formation; RSA: “Rosso Ammonitico” formation; MAS: “Calcare Massiccio” formation; RET: “Raethavicola Contorta” formation, Late Triassic dolomitic-limestones; ctc: cataclastic formation made by a polygenic breccias of mainly metamorphic limestone clasts. Tuscan metamorphic units: PSM: “Pseudomacigno” formation; MCP: “Cipollini” formation. microsamples of dentin and bones do (Figures 10(e) and been more exposed to postmortem diagenetic exchange 10(f)). )is strongly suggests that 87Sr/86Sr of tooth enamel is processes compared to more heavily mineralized enamel. )is unaffected by diagenetic alteration, in contrast to the other speaks for a possible isotope reequilibration between dentin organic-rich samples (i.e., bones and dentin). )e 87Sr/86Sr and bones (but not enamel) and soil, due to Sr exchange with composition of the dentin samples is close to that of the local percolating fluids. )e soil samples have 87Sr/86Sr values soil samples (Figure 10(e)); even closer to the latter is that of comparable to those of the “Calcare a Rhaetavicula Contorta” the bone samples, with the only exception of SCT3 due to its formation, which forms the cave’s bedrock [129], through high degree of mineralisation. which fluids filter into the cave (Figure 10(e)). )ese results, despite the relatively recent age of the fossil In summary, our study shows that the Grotta specimens (ca. 40 ka [127]), indicate that bone tissues have all’Onda bones and dentin are unsuitable to determine the 14 Journal of Analytical Methods in Chemistry

Tooth sample Bone sample Bone sample SCT-6 1mm SCT-2 SCT-3 Marrow Marrow Dentin

Bone Enamel Dentin Bone 1cm 1cm 1cm Enamel

(a) (b) (c) (d) 0.7092

0.7090 Dolomitic 0.7092 fraction Enamel 0.7088 SCT3

Sr 0.7088

86 0.7086 Dentin

Sr/ 86

87 SCT2

0.7084 Sr/ 0.7084 SCT1 87 Carbonatic fraction Rhaetavicola contorta limestone 0.7082 0.7080 Rhaetavicola contorta limestone 0.7080 3.4 3.3 3.2 3.1 3.0 2.9 2.8 – – – – – – – 0.7076 log Sr/Ca –3.5 –3.4 –3.3 –3.2 –3.1 –3.0 –2.9 –2.8 –2.7 –2.6 –2.5 log Sr/Ca Bones SCT6 enamel SCT5 dentin SCT4 enamel Soils SCT6 dentin SCT4 enamel SCT6 enamel SCT5 enamel SCT4 dentin SCT5 enamel

(e) (f)

Figure 10: Results of micro-Sr isotopes in teeth and bones of the Ursus spelaeus. (a) Representative image of one tooth sample, sectioned for microdrilling. )e thin layer of enamel is well evident with respect to the lighter inner dentin. In yellow are reported the two drilling sites on dentin and enamel; (b) particular back-scattered electron microscopy image of the edge of the tooth showing dentin (dark grey) and enamel (light grey) portion; (c and d) images of two of the analysed bones properly prepared in epoxy resin mounts for microdrilling sampling. In yellow is reported the drilling site; (e) 87Sr/86Sr versus Sr/Ca (in log scale) diagram showing the results obtained from the analyses of enamel and dentin in the three teeth, compared with bones of the Ursus spelaeus specimen. )e 87Sr/86Sr of soils in which the fossils have been sampled are also reported together with the Sr-isotope field of the Raetavicola Contorta limestone for comparison. (See text for detail) (f) 87Sr/86Sr versus Sr/Ca (in log scale) diagram comparing the Sr-isotope values of dentin from the three teeth samples with the Sr-isotope range fields of the different geologic formations forming the bedrock outcropping in the area of the Grotta all’Onda cave.

Table 4: (a) 87Sr/86Sr results obtained on microsamples of tooth dentine and enamel and bones of the Ursus spelaeus specimen. (b) 87Sr/86Sr soils sampled in the Grotta all’Onda cave from which the fossil remains were collected. Log (Sr/Ca) Sample Type Sr content (ppm) 87Sr/86Sr 2 SE Average (a) Teeth and bones STC1 Bone 510 0.708268 0.000006 −2.84 STC2 Bone 352 0.708441 0.000006 −2.98 STC3 Bone 359 0.708758 0.000006 −2.95 STC4-1 Enamel 187 0.708703 0.000005 −3.32 STC4-2 Dentin 360 0.708362 0.000006 −3.00 STC5-1 Enamel 211 0.708663 0.000005 −3.21 STC5-2 Dentin 384 0.708380 0.000006 −2.94 STC6-1 Enamel 179 0.709081 0.000005 −3.30 STC6-2 Dentin 339 0.708504 0.000006 −3.02 (b) Soils STC7 Soil 204 0.708184 0.000006 −2.86 STC8 Soil 228 0.708155 0.000006 −2.86 2 SE: two standard error of the mean. characteristics of the habitat where the cave bear lived, due to In contrast, the 87Sr/86Sr ratios of the tooth enamel results their interaction with percolating water and their consequent unaltered and realistically reflects the original values achieved contamination by the soils in which they had been preserved. during the cave bear’s life. )e isotopic composition of the Journal of Analytical Methods in Chemistry 15 enamel samples is consistent with that of the “Calcare Conflicts of Interest Massiccio Formation” and in particular with the dolomite fraction [129] (Figure 10(f)). )e mismatch between the )e authors declare that there are no conflicts of interest 87Sr/86Sr values of the enamel samples and those of the cave regarding the publication of this paper. soils (Figure 10(e)) (i.e., local substrata of the cave) indicate that the cave bear died away from its customary habitat. Bears Authors’ Contributions cannot indeed find food in caves, where they find refuge as shelter for winter hibernation or for night resting. )e enamel Lorella Francalanci provided the input to the set-up and isotopic values obtained during our study indicate that Ursus the development of the in situ micro-Sr isotope procedure spelaeus from Grotta all’Onda roamed in search for food at the Radiogenic Isotope Laboratory of Firenze, providing within a confined area not far from the cave, where the also continuous scientific stirring and encouraging to the “Calcare Massiccio” is largely exposed and did not move too other research fellows, coauthors of the present paper. far from the area during its whole life. Acknowledgments 4. Summary and Conclusions )e research on volcanic rocks was financially supported by )e present study show the potential of 87Sr/86Sr de- PRIN 2010-2011 and 2015 with Grants 2010TT22SC_001 and termination by TIMS on micro-scale samples, based on 20158A9CBM, respectively. Regione Toscana government micromilling solid specimens, not only for geological ap- supported the salary of Sara Di Salvo through a “Pegaso” plications, but also for other fields, such as archaeology, fellowship. )e studies on the Ursus spelaeus had no financial forensics, medical, and life sciences, where it has hardly, if support, and the analytical work expenses were covered by the ever, been used. Reported here is a detailed description of all “Radiogenic Lab” of the University of Florence. )e authors the analytical protocols, including results on replicate an- are grateful for the suggestions of an anonymous reviewer, alyses of international standards (SRM 987 and BHVO-2G), which improved the paper. which yield good accuracy and precision. In addition, three case studies are presented, performed in our laboratory, References where in situ microdrilled Sr isotopes have been used in different fields of application. [1] F. J. Tepley and J. P. Davidson, “Mineral-scale Sr-isotope )e first case study on micro-Sr isotope determination at constraints on magma evolution and chamber dynamics in subgrain microscopic scale regards the petrogenetic pro- the Rum layered intrusion, Scotland,” Contributions to cesses relevant to the understanding of the plumbing system Mineralogy and Petrology, vol. 145, no. 5, pp. 628–641, 2003. dynamics under active volcanoes. )is example revealed the [2] G. Perini III, F. J. Tepley, J. P. Davidson, and S. Conticelli, “)e origin of K-feldspar megacrysts hosted in alkaline role played by the interaction of different magmas, which are 87 86 potassic rocks from central Italy: a track for low-pressure normally characterised by distinct Sr/ Sr signatures, processes in mafic magmas,” Lithos, vol. 66, no. 3-4, comingled in the plumbing system of the Nisyros volcano, pp. 223–240, 2003. which was capable of triggering the eruption. [3] D. Morgan, S. Blake, N. Rogers et al., “Time scales of crystal )e second case regards volcanic glasses with extremely residence and magma chamber volume from modelling of low total Sr content (i.e., tholeiitic). )e micromilling de- diffusion profiles in phenocrysts: vesuvius 1944,” Earth and termination of 87Sr/86Sr ratio was performed on ashes, with Planetary Science Letters, vol. 222, no. 3-4, pp. 933–946, different shapes and nature, erupted by different phases, 2004. during the 2010 eruption of Eyjafjallajokull¨ volcano (Ice- [4] W. Siebel, E. Reitter, T. Wenzel, and U. Blaha, “Sr isotope land). )e 87Sr/86Sr data provide information on the eruptive systematics of K-feldspars in plutonic rocks revealed by the mechanism involved during the eruption, as well as on the Rb-Sr microdrilling technique,” Chemical Geology, vol. 222, no. 3-4, pp. 183–199, 2005. interaction between the magma and the hydrothermally- [5] L. Francalanci, G. R. Davies, W. Lustenhouwer, derived minerals attained before the thawing of the ice cap; it S. Tommasini, P. R. Mason, and S. Conticelli, “Intra-grain Sr also provides significant insights into the interpretation of isotope evidence for crystal recycling and multiple magma the precursory signals of the eruption. reservoirs in the recent activity of Stromboli volcano, )e third case displays the use of 87Sr/86Sr microdrilled southern Italy,” Journal of Petrology, vol. 46, no. 10, in enamel, dentin, and bones, to show that only enamel has pp. 1997–2021, 2005. more chances to preserve the original Sr-isotope signatures [6] L. Francalanci, R. Avanzinelli, I. Nardini, M. Tiepolo, than bone and dentin. )e analysis also revealed the close J. P. Davidson, and R. Vannucci, “Crystal recycling in the relationship existing between the radiogenic-Sr of the or- steady-state system of the active Stromboli volcano: a 2.5-ka ganic materials and that of the geologic Cretaceous sub- story inferred from in situ Sr-isotope and trace element data,” Contributions to Mineralogy and Petrology, vol. 163, stratum of the Apuan Alps, which provides valuable insights no. 1, pp. 109–131, 2012. into the palaeoenvironment of the local cave bears. In this case, [7] J. P. Davidson, D. J. Morgan, B. L. A. Charlier, R. Harlou, and the Sr isotopes proved particularly useful for determining the J. M. Hora, “Microsampling and isotopic analysis of igneous foraging habits of extinct mammals, which substantiates the rocks: implications for the study of magmatic systems,” well-known statement “YOU ARE WHAT YOU EAT” (cit. Annual Review of Earth and Planetary Sciences, vol. 35, no. 1, Anthelme Brillat-Savarin). pp. 273–311, 2007. 16 Journal of Analytical Methods in Chemistry

[8] E. Braschi, L. Francalanci, S. Tommasini, and G. E. Vougioukalakis, era and the present from Pb and Sr isotopes,” Science of the “Unraveling the hidden origin and migration of plagioclase Total Environment, vol. 224, no. 1, pp. 109–119, 1998. phenocrysts by in situ Sr isotopes: the case of final dome activity [22] R. C. Capo, B. W. Stewart, and O. A. Chadwick, “Strontium at Nisyros volcano, Greece,” Contributions to Mineralogy and isotopes as tracers of ecosystem processes: theory and Petrology, vol. 167, no. 3, p. 988, 2014. methods,” Geoderma, vol. 82, no. 1–3, pp. 197–225, 1998. [9] S. Conticelli, E. Boari, L. Burlamacchi et al., “Geochemistry [23] B. L. Beard and C. M. Johnson, “Strontium isotope com- and Sr-Nd-Pb isotopes of Monte Amiata Volcano, Central position of skeletal material can determine the birth place Italy: evidence for magma mixing between high-K calc- and geographic mobility of humans and animals,” Journal of alkaline and leucititic mantle-derived magmas,” Italian Journal Forensic Science, vol. 45, no. 5, pp. 1049–1061, 2000. of Geosciences, vol. 134, no. 2, pp. 266–290, 2015. [24] S. Tommasini, G. R. Davies, and T. Elliott, “Lead isotope [10] M. Pistolesi, R. Cioni, L. Francalanci et al., “)e onset of an composition of tree rings as bio-geochemical tracers of heavy eruption: selective assimilation of hydrothermal minerals metal pollution: a reconnaissance study from Firenze, Italy,” during pre-eruptive magma ascent of the 2010 summit Applied Geochemistry, vol. 15, no. 7, pp. 891–900, 2000. eruption of Eyjafjallajokull¨ volcano, Iceland,” Journal of [25] M. Vanhaeren, F. d’Errico, I. Billy, and F. Grousset, “Tracing Volcanology and Geothermal Research, vol. 327, pp. 449–458, the source of Upper Palaeolithic shell beads by strontium 2016. isotope dating,” Journal of Archaeological Science, vol. 31, [11] J. N. Christensen, A. N. Halliday, D. C. Lee, and C. M. Hall, no. 10, pp. 1481–1488, 2004. “In situ Sr isotopic analysis by laser ablation,” Earth and [26] C. M. R. Almeida and M. T. S. Vasconcelos, “Lead contam- Planetary Science Letters, vol. 136, no. 1-2, pp. 79–85, 1995. ination in Portuguese red wines from the Douro region: from [12] J. P. Davidson and F. J. Tepley, “Recharge in volcanic sys- the vineyard to the final product,” Journal of Agricultural and tems: evidence from isotope profiles of phenocrysts,” Science, Food Chemistry, vol. 51, no. 10, pp. 3012–3023, 2003. vol. 275, no. 5301, pp. 826–829, 1997. [27] J. A. Evans and S. Tatham, “Defining “local signature” in [13] J. P. Davidson, L. Font, B. L. Charlier, and F. J. Tepley, terms of Sr isotope composition using a tenth-to twelfth- “Mineral-scale Sr isotope variation in plutonic rocks–a tool century Anglo-Saxon population living on a Jurassic clay- for unravelling the evolution of magma systems,” Earth and carbonate terrain, Rutland, UK,” Geological Society, London, Environmental Science Transactions of the Royal Society of Special Publications, vol. 232, no. 1, pp. 237–248, 2004. Edinburgh, vol. 97, no. 4, pp. 357–367, 2006. [28] S. Holzl,¨ P. Horn, A. Rossmann, and S. Rummel, “Isotope- [14] B. L. A. Charlier, O. Bachmann, J. P. Davidson, abundance ratios of light (bio) and heavy (geo) elements in M. A. Dungan, and D. J. Morgan, “)e upper crustal evo- biogenic tissues: methods and applications,” Analytical and lution of a large silicic magma body: evidence from crystal- Bioanalytical Chemistry, vol. 378, no. 2, pp. 270–272, 2004. scale Rb–Sr isotopic heterogeneities in the Fish Canyon [29] S. Kelly, K. Heaton, and J. Hoogewerff, “Tracing the geo- magmatic system, Colorado,” Journal of Petrology, vol. 48, graphical origin of food: the application of multi-element no. 10, pp. 1875–1894, 2007. and multi-isotope analysis,” Trends in Food Science & [15] B. L. Charlier, C. J. Wilson, and J. P. Davidson, “Rapid open- Technology, vol. 16, no. 12, pp. 555–567, 2005. system assembly of a large silicic magma body: time-resolved [30] R. A. Bentley and C. Knipper, “Geographical patterns in evidence from cored plagioclase crystals in the Oruanui biologically available strontium, carbon and oxygen isotope eruption deposits, New Zealand,” Contributions to Miner- signatures in prehistoric SW Germany,” Archaeometry, alogy and Petrology, vol. 156, no. 6, pp. 799–813, 2008. vol. 47, no. 3, pp. 629–644, 2005. [16] J. P. Davidson, F. J. Tepley, Z. Palacz, and S. Meffan-Main, [31] P. Degryse, J. Schneider, U. Haack et al., “Evidence for glass “Magma recharge, contamination, and residence times “recycling” using Pb and Sr isotopic ratios and Sr-mixing revealed by in situ laser ablation isotopic analysis of feldspar lines: the case of early Byzantine Sagalassos,” Journal of in volcanic rocks,” Earth and Planetary Science Letters, Archaeological Science, vol. 33, no. 4, pp. 494–501, 2006. vol. 184, no. 2, pp. 427–442, 2001. [32] S. Rummel, S. Hoelzl, P. Horn, A. Rossmann, and C. Schlicht, [17] F. J. Tepley, J. P. Davidson, R. I. Tilling, and J. G. Arth, “)e combination of stable isotope abundance ratios of H, C, “Magma mixing, recharge and eruption histories recorded in N and S with 87Sr/86Sr for geographical origin assignment of plagioclase phenocrysts from El Chichon Volcano, Mexico,” orange juices,” Food Chemistry, vol. 118, no. 4, pp. 890–900, Journal of Petrology, vol. 41, no. 9, pp. 1397–1411, 2000. 2010. [18] D. J. Morgan, D. A. Jerram, D. G. Chertkoff et al., “Com- [33] S. Vorkelius, G. D. Lorenz, S. Rummel et al., “Strontium bining CSD and isotopic microanalysis: magma supply and isotopic signatures of natural mineral waters, the reference mixing processes at Stromboli Volcano, Aeolian Islands, to a simple geological map and its potential for authenti- Italy,” Earth and Planetary Science Letters, vol. 260, no. 3-4, cation of food,” Food Chemistry, vol. 118, no. 4, pp. 933– pp. 419–431, 2007. 940, 2010. [19] P. Horn, P. Schaaf, B. Holbach, S. Holzl,¨ and H. Eschnauer, [34] B. P. Kennedy, A. Klaue, J. D. Blum, C. L. Folt, and “87Sr/86Sr from rock and soil into vine and wine,” Zeitschrift K. H. Nislow, “Reconstructing the lives of fish using Sr für Lebensmitteluntersuchung und-Forschung A, vol. 196, isotopes in otoliths,” Canadian Journal of Fisheries and no. 5, pp. 407–409, 1993. Aquatic Sciences, vol. 59, no. 6, pp. 925–929, 2011. [20] J. L. Banner and J. Kaufman, “)e isotopic record of ocean [35] C. Durante, C. Baschieri, L. Bertacchini et al., “Geographical chemistry and diagenesis preserved in non-luminescent traceability based on 87Sr/86Sr indicator: a first approach for brachiopods from Mississippian carbonate rocks, Illinois and PDO Lambrusco wines from Modena,” Food Chemistry, Missouri,” Geological Society of America Bulletin, vol. 106, vol. 141, no. 3, pp. 2779–2787, 2013. no. 8, pp. 1074–1082, 1994. [36] S. Marchionni, E. Braschi, S. Tommasini et al., “High-pre- [21] G. Aberg,˚ G. Fosse, and H. Stray, “Man, nutrition and cision 87Sr/86Sr analyses in wines and their use as a geological mobility: a comparison of teeth and bone from the Medieval fingerprint for tracing geographic provenance,” Journal of Journal of Analytical Methods in Chemistry 17

Agricultural and Food Chemistry, vol. 61, no. 28, pp. 6822– [51] S. R. Copeland, M. Sponheimer, P. J. le Roux et al., 6831, 2013. “Strontium isotope ratios (87Sr/86Sr) of tooth enamel: [37] S. Marchionni, A. Buccianti, A. Bollati et al., “Conservation a comparison of solution and laser ablation multicollector of 87Sr/86Sr isotopic ratios during the winemaking processes inductively coupled plasma mass spectrometry methods,” of “Red” wines to validate their use as geographic tracer,” Rapid Communications in Mass Spectrometry, vol. 22, no. 20, Food Chemistry, vol. 190, pp. 777–785, 2016. pp. 3187–3194, 2008. [38] V. Vinciguerra, R. Stevenson, K. Pedneault, A. Poirier, [52] A. Simonetti, M. R. Buzon, and R. A. Creaser, “In situ el- J.-F. Helie,´ and D. Widory, “Strontium isotope character- emental and Sr isotope investigation of human tooth enamel ization of wines from Quebec, Canada,” Food Chemistry, by Laser Ablation-(MC)-ICP-MS: success and pitfalls,” vol. 210, pp. 121–128, 2016. Archaeometry, vol. 50, no. 2, pp. 371–385, 2008. [39] P. L. Koch, J. Heisinger, C. Moss, R. W. Carlson, M. L. Fogel, [53] J. Lewis, C. D. Coath, and A. W. G. Pike, “An improved and A. K. Behrensmeyer, “Isotopic tracking of change in diet protocol for 87Sr/86Sr by laser ablation multi-collector in- and habitat use in African elephants,” Science, vol. 267, ductively coupled plasma mass spectrometry using oxide no. 5202, p. 1340, 1995. reduction and a customised plasma interface,” Chemical [40] A. Sillen, G. Hall, and R. Armstrong, “Strontium calcium Geology, vol. 390, pp. 173–181, 2014. ratios (Sr/Ca) and strontium isotopic ratios (87Sr/86Sr) of [54] B. L. A. Charlier, C. Gimbre, D. Morgan et al., “Methods of Australopithecus robustus and Homo sp. From Swartkrans,” microsampling and high-precision analysis of strontium and Journal of Human Evolution, vol. 28, no. 3, pp. 277–285, 1995. rubidium isotopes at single crystal scale for petrological and [41] G. Grupe, “Preservation of collagen in bone from dry, Sandy geochronological applications,” Chemical Geology, vol. 232, soil,” Journal of Archaeological Science, vol. 22, no. 2, no. 3-4, pp. 144–133, 2006. pp. 193–199, 1995. [55] R. Avanzinelli, E. Boari, S. Conticelli et al., “High precision [42] G. Grupe, T. D. Price, and F. Söllner,“Mobility of Bell Beaker Sr, Nd, and Pb isotopic analyses using the new generation people revealed by strontium isotope ratios of tooth and )ermal Ionisation Mass Spectrometer )ermoFinningan bone: a study of southern Bavarian skeletal remains. A reply Triton-Ti®,” Periodico di Mineralogia, vol. 74, no. 3, to the comment by Peter Horn and Dieter Müller-Sohnius,” pp. 147–166, 2005. Applied Geochemistry, vol. 14, no. 2, pp. 271–275, 1999. [56] D. L. Hoffmann, C. Spötl,and A. Mangini, “Micromill and in [43] P. Willey, A. Galloway, and L. Snyder, “Bone mineral situ laser ablation sampling techniques for high spatial density and survival of elements and element portions in resolution MC-ICPMS U-) dating of carbonates,” Chem- the bones of the Crow Creek massacre victims,” American ical Geology, vol. 259, no. 3-4, pp. 253–261, 2009. Journal of Physical Anthropology, vol. 104, no. 4, pp. 513– [57] A. Eberhart, R. Delwiche, and Z. Geiss, “Isotopic effects in 528, 1997. single filament thermal ion sources,” Zeitschrift fürNatur- [44] K. A. Hoppe, P. L. Koch, R. W. Carlson, and S. D. Webb, forschung A, vol. 19, no. 6, pp. 736–740, 1964. “Tracking mammoths and mastodons: reconstruction of [58] W. A. Russel, D. A. Papanastassiou, and T. A. Tombrello, “Ca migratory behavior using strontium isotope ratios,” Geology, isotope fractionation on the Earth and other solar system vol. 27, no. 5, pp. 439–442, 1999. materials,” Geochimica et Cosmochimica Acta, vol. 42, no. 8, [45] T. D. Price, J. H. Burton, and R. A. Bentley, “)e charac- pp. 1075–1090, 1978. terization of biologically available strontium isotope ratios [59] M. F. )irlwall, “Long-term reproducibility of multicollector for the study of prehistoric migration,” Archaeometry, Sr and Nd isotope ratio analysis,” Chemical Geology, vol. 94, vol. 44, no. 1, pp. 117–135, 2002. no. 2, pp. 85–104, 1991. [46] R. A. Bentley, T. D. Price, and E. Stephan, “Determining the [60] D. Weis, B. Kieffer, C. Maerschalk et al., “High-precision “local” 87Sr/86Sr range for archaeological skeletons: a case isotopic characterization of USGS reference materials by study from Neolithic Europe,” Journal of Archaeological TIMS and MC-ICP-MS,” Geochemistry, Geophysics, Geo- Science, vol. 31, no. 4, pp. 365–375, 2004. systems, vol. 7, no. 8, 2006. [47] J. A. Evans, J. Montgomery, G. Wildman, and N. Boulton, [61] J. B. Mahoney, D. Weiss, B. Keiffer et al., “Ongoing isotopic “Spatial variations in biosphere 87Sr/86Sr in Britain,” characterization of USGS standards: MC-ICPMS and TIMS Journal of the Geological Society, vol. 167, no. 1, pp. 1–4, data from the Pacific Centre for isotopic and geochemical 2010. research, University of British Columbia,” in Proceedings of [48] J. H. Burton and T. D. Price, “Seeking the local 87Sr/86Sr ratio Geological Society of America, Seattle Annual Meeting, Paper to determine geographic origins of humans,” in ACS Sym- 117-118, Seattle, WA, USA, October 2003. posium Series, pp. 309–320, ACS Publication, Washington, [62] M. Elburg, P. Vroon, B. van der Wagt, and A. Tchalikian, “Sr DC, USA, 2013. and Pb isotopic composition of five USGS glasses (BHVO- [49] F. Lugli, A. Cipriani, C. Peretto, M. Mazzucchelli, and 2G, BIR-1G, BCR-2G, TB-1G, NKT-1G),” Chemical Geology, D. Brunelli, “In situ high spatial resolution 87Sr/86Sr ratio vol. 223, no. 4, pp. 196–207, 2005. determination of two Middle Pleistocene (ca 580 ka) Ste- [63] C. Spotl¨ and D. Mattey, “Stable isotope microsampling of phanorhinus hundsheimensis teeth by LA–MC–ICP–MS,” speleothems for palaeoenvironmental studies: a comparison International Journal of Mass Spectrometry, vol. 412, of microdrill, micromill and laser ablation techniques,” pp. 38–48, 2017. Chemical Geology, vol. 235, no. 1-2, pp. 48–58, 2006. [50] T. Prohaska, C. Latkoczy, G. Schultheis, M. Teschler- [64] C. Saenger, R. I. Gabitov, J. Farmer, J. M. Watkins, and Nicola, and G. Stingeder, “Investigation of Sr isotope ratios R. Stone, “Linear correlations in bamboo coral δ13C and δ18O in prehistoric human bones and teeth using laser ablation sampled by SIMS and micromill: evaluating paleoceano- ICP-MS and ICP-MS after Rb/Sr separation,” Journal of graphic potential and biomineralization mechanisms using 11 Analytical Atomic Spectrometry, vol. 17, no. 8, pp. 887–891, δ B and ∆47 composition,” Chemical Geology, vol. 454, 2002. pp. 1–14, 2017. 18 Journal of Analytical Methods in Chemistry

[65] K. M. Knesel, J. P. Davidson, and W. A. Duffield, “Evolution Akaki Canyon, Cyprus: implications for the origin of the of silicic magma through assimilation and subsequent re- Troodos ophiolite,” Earth and Planetary Science Letters, charge: evidence from Sr isotopes in sanidine phenocrysts, vol. 75, no. 4, pp. 369–383, 1985. Taylor Creek Rhyolite, NM,” Journal of Petrology, vol. 40, [80] N. Blum and J. H. Crocket, “Repetitive cyclical volcanism no. 5, pp. 773–786, 1999. in the Late Archean Larder Lake Group near Kirkland [66] G. S. Wallace and G. W. Bergantz, “Reconciling heteroge- Lake, Ontario: implications of geochemistry on magma neity in crystal zoning data: an application of shared char- genesis,” Precambrian Research, vol. 54, no. 2-4, pp. 173– acteristic diagrams at Chaos Crags, Lassen Volcanic Center, 194, 1992. California,” Contributions to Mineralogy and Petrology, [81] M. F. Roden, T. Trull, S. R. Hart, and F. A. Frey, “New He, vol. 149, no. 1, pp. 98–112, 2005. Nd, Pb, and Sr isotopic constraints on the constitution of the [67] O. Bachmann and G. W. Bergantz, “Gas percolation in Hawaiian plume: results from Koolau Volcano, Oahu, upper-crustal silicic crystal mushes as a mechanism for Hawaii, USA,” Geochimica et Cosmochimica Acta, vol. 58, upward heat advection and rejuvenation of near-solidus no. 5, pp. 1431–1440, 1994. magma bodies,” Journal of Volcanology and Geothermal [82] P. R. Castillo, E. Klein, J. Bender et al., “Petrology and Sr, Nd, Research, vol. 149, no. 1-2, pp. 85–102, 2006. and Pb isotope geochemistry of mid-ocean ridge basalt [68] L. Font, J. P. Davidson, D. G. Pearson, G. M. Nowell, glasses from the 11 45′ N to 15 00′ N segment of the East D. A. Jerram, and C. J. Ottley, “Sr and Pb isotope micro- Pacific Rise,” Geochemistry, Geophysics, Geosystems, vol. 1, analysis of plagioclase crystals from Skye lavas: an insight no. 11, 2000. into open-system processes in a flood basalt province,” [83] L. Melluso, V. Morra, P. Brotzu et al., “Geochronology and Journal of Petrology, vol. 49, no. 8, pp. 1449–1471, 2008. petrogenesis of the Cretaceous Antampombato–Ambatovy [69] V. M. Martin, J. Davidson, D. Morgan, and D. A. Jerram, complex and associated dyke swarm, Madagascar,” Journal “Using the Sr isotope compositions of feldspars and glass to of Petrology, vol. 46, no. 10, pp. 1963–1996, 2005. distinguish magma system components and dynamics,” [84] G. Conde, P. D. Ihinger, and E. E. Frahm, “Water speciation Geology, vol. 38, no. 6, pp. 539–542, 2010. in Anatolian Obsidian: quenched magmatic water vs low [70] G. M. Di Paola, “Volcanology and petrology of Nisyros is- temperature hydration,” Geochimica et Cosmochimica Acta, land (Dodecanese Greece),” Bulletin Volcanologique, vol. 38, vol. 73, p. A239, 2009. no. 3, pp. 944–987, 1974. [85] P. Degryse, A. Boyce, N. E. Satullo et al., “Isotopic dis- [71] G. P. Wyers and M. Barton, “Polybaric evolution of criminants between late Bronze Age glasses from Egypt and calc-alkaline magmas from Nisyros, southeastern Hellenic the Near East,” Archaeometry, vol. 52, no. 3, pp. 380–388, Arc, Greece,” Journal of Petrology, vol. 30, no. 1, pp. 1–37, 2010. 1989. [86] J. Henderson, J. Evans, and K. Nikita, “Isotopic evidence for [72] K. S. Seymour and D. Vlassopoulos, “Magma mixing at the primary production, provenance and trade of Late Nisyros volcano, as inferred from incompatible trace-ele- Bronze Age glass in the Mediterranean,” Mediterranean ment systematics,” Journal of Volcanology and Geothermal Archaeology and Archaeometry, vol. 10, no. 1, pp. 1–24, 2010. Research, vol. 50, no. 3, pp. 273–299, 1992. [87] M. Ganio, K. Latruwe, D. Brems, P. Muchez, F. Vanhaecke, [73] G. Vougioukalakis, “Volcanic stratigraphy and evolution of and P. Degryse, “)e Sr–Nd isolation procedure for sub- Nisyros island,” Bulletin of the Geological Society of Greece, sequent isotopic analysis using multi-collector ICP-mass vol. 28, no. 2, pp. 239–258, 1993. spectrometry in the context of provenance studies on ar- [74] L. Francalanci, J. C. Varekamp, E. G. Vougioukalakis, chaeological glass,” Journal of Analytical Atomic Spectrom- M. J. Defant, F. Innocenti, and P. Manetti, “Crystal retention, etry, vol. 27, no. 8, pp. 1335–1341, 2012. fractionation and crustal assimilation in a convecting magma [88] E. Gliozzo, E. Braschi, F. Giannetti, A. Langone, and chamber, Nisyros Voclano, Greece,” Bulletin of Volcanology, M. Turchiano, “New geochemical and isotopic insights vol. 56, no. 8, pp. 601–620, 1995. into the Late Antique Apulian glass and the HIMT1 and [75] L. Francalanci, J. C. Varekamp, E. G. Vougioukalakis, HIMT2 glass productions—the glass vessels from San F. Innocenti, and P. Manetti, “Is there a compositional gap at Giusto (Foggia, Italy) and the diagrams for provenance Nisyros volcano? A comment on: magma generation at the studies,” Archaeological and Anthropological Sciences, easternmost section of the Hellenic arc: Hf, Nd, Pb and Sr pp. 1–30, 2017. isotope geochemistry of Nisyros and Yali volcanoes [89] F. Sigmundsson, S. Hreinsdóttir,A. Hooper et al., “Intrusion (Greece),” Lithos, vol. 95, no. 3-4, pp. 458–461, 2007. triggering of the 2010 Eyjafjallajokull explosive eruption,” [76] C. Longchamp, A. Skopelitis, C. Bonadonna, and Nature, vol. 468, no. 7322, pp. 426–430, 2010. O. Bachmann, “Characterization of tephra deposits with [90] E. Kaminski, S. Tait, F. Ferrucci, M. Martet, B. Hirn, and limited exposure: the example of the two largest explosive P. Husson, “Estimation of ash injection in the atmosphere by eruptions at Nisyros volcano (Greece),” Bulletin of Volca- basaltic volcanic plumes: the case of the Eyjafjallajokull¨ 2010 nology, vol. 73, no. 9, pp. 1337–1352, 2011. eruption,” Journal of Geophysical Research: Solid Earth, [77] E. M. Limburg and J. C. Varekamp, “Young pumice deposits vol. 116, no. 9, 2011. on Nisyros, Greece,” Bulletin of Volcanology, vol. 54, no. 1, [91] M. T. Gudmundsson, T. )ordarson, A. Höskuldssonet al., pp. 68–77, 1991. “Ash generation and distribution from the April-May 2010 [78] E. Braschi, L. Francalanci, and G. E. Vougioukalakis, “In- eruption of Eyjafjallajokull,¨ Iceland,” Scientific Reports, vol. 2, verse differentiation pathway by multiple mafic magma no. 1, p. 572, 2012. refilling in the last magmatic activity of Nisyros Volcano, [92] P. W. Webley, T. Steensen, M. Stuefer, G. Grell, S. Freita, and Greece,” Bulletin of Volcanology, vol. 74, no. 5, pp. 1083– M. Pavolonis, “Analyzing the Eyjafjallajökull2010 eruption 1100, 2012. using satellite remote sensing, lidar and WRF-Chem dis- [79] M. Rautenschlein, G. A. Jenner, J. Hertogen et al., “Isotopic persion and tracking model,” Journal of Geophysical Re- and trace element composition of volcanic glasses from the search: Atmospheres, vol. 117, no. 20, 2012. Journal of Analytical Methods in Chemistry 19

[93] R. Cioni, M. Pistolesi, A. Bertagnini, C. Bonadonna, [109] J. E. Ericson, “Some problems and potentials of strontium A. Hoskuldsson, and B. Scateni, “Insights into the dynamics isotope analysis for human and animal ecology,” in Stable and evolution of the 2010 Eyjafjallajökullsummit eruption Isotopes in Ecological Research, pp. 252–259, Springer, New (Iceland) provided by volcanic ash textures,” Earth and York, NY, USA, 1989. Planetary Science Letters, vol. 394, pp. 111–123, 2014. [110] J. C. Sealy, N. J. van der Merwe, A. Sillen, F. J. Kruger, and [94] O. Sigmarsson, J. Maclennan, and M. Carpentier, “Geo- H. W. Krueger, “87Sr86Sr as a dietary indicator in modern chemistry of igneous rocks in Iceland: a review,” Jokull¨ , and archaeological bone,” Journal of Archaeological Science, vol. 58, pp. 139–160, 2008. vol. 18, no. 3, pp. 399–416, 1991. [95] T. D. Price, S. Nakamura, S. Suzuki, J. H. Burton, and [111] B. K. Nelson, M. J. Deniro, M. Schoeninger, D. J. De Paolo, V. Tiesler, “New isotope data on Maya mobility and enclaves and P. E. Hare, “Effects of diagenesis on strontium, carbon, at Classic Copan, Honduras,” Journal of Anthropological nitrogen and oxygen concentration and isotopic composi- Archaeology, vol. 36, pp. 32–47, 2014. tion of bone,” Geochimica et Cosmochimica Acta, vol. 50, [96] P. A. Slater, K. M. Hedman, and T. E. Emerson, “Immigrants no. 9, pp. 1941–1949, 1986. at the Mississippian polity of Cahokia: strontium isotope [112] P. Budda, B. L. J. Montgomery, P. Rainbird, R. G. )omas, evidence for population movement,” Journal of Archaeo- and S. M. Young, “Pb and Sr isotope composition of human logical Science, vol. 44, pp. 117–127, 2014. dental enamel: an indicator of Pacific Islander population [97] A. J. Waterman, D. W. Peate, A. M. Silva, and J. T. )omas, dynamics,” in Le Pacifique de 5000 a` 2000 avant le present:´ “In search of homelands: using strontium isotopes to supplements´ a` l’histoire d’une colonisation � +e Pacific from identify biological markers of mobility in late prehistoric 5000 to 2000 BP: Colonisation and Transformations, G. Jean- Portugal,” Journal of Archaeological Science, vol. 42, Christophe and I. Lilley, Eds., Institut de recherche pour le pp. 119–127, 2014. developpement,´ Marseille, France, 1999. [98] A. L. Rheingold, S. Hues, and M. N. Cohen, “Strontium and [113] M. Sponheimer and J. A. Lee-)orp, “Enamel diagenesis at zinc content in bone as an indication of diet,” Journal of South African Australopith sites: implications for paleo- Chemical Education, vol. 60, no. 3, pp. 233-234, 1983. ecological reconstruction with trace elements,” Geochimica [99] C. Gilbert, J. Sealy, and A. Sillen, “An investigation of et Cosmochimica Acta, vol. 70, no. 7, pp. 1644–1654, 2006. barium, calcium and strontium as palaeodietary indicators in [114] C. Hanni,¨ V. Laudet, D. Stehelin, and P. Taberlet, “Tracking the Southwestern Cape, South Africa,” Journal of Archaeo- the origins of the cave bear (Ursus spelaeus) by mitochondrial logical Science, vol. 21, no. 2, pp. 173–184, 1994. DNA sequencing,” Proceedings of the National Academy of [100] T. D. Price, C. M. Johnson, J. A. Ezzo, J. Ericson, and Sciences, vol. 91, no. 25, pp. 12336–12340, 1994. J. H. Burton, “Residential mobility in the prehistoric [115] P. Argenti and P. P. Mazza, “Mortality analysis of the Late southwest United States: a preliminary study using strontium Pleistocene bears from Grotta Lattaia, central Italy,” Journal isotope analysis,” Journal of Archaeological Science, vol. 21, of Archaeological Science, vol. 33, no. 11, pp.1552–1558, 2006. no. 3, pp. 315–330, 1994. [116] I. Martini, M. Coltorti, P. P. Mazza, M. Rustioni, and [101] T. D. Price, L. Manzanilla, and W. D. Middleton, “Immi- F. Sandrelli, “)e latest Ursus spelaeus in Italy, a new con- gration and the ancient city of Teotihuacan in Mexico: tribution to the extinction chronology of the cave bear,” a study using strontium isotope ratios in human bone and Quaternary Research, vol. 81, no. 1, pp. 117–124, 2014. teeth,” Journal of Archaeological Science, vol. 27, no. 10, [117] H. Bocherens, M. Fizet, and A. Mariotti, “Diet, physiology pp. 903–913, 2000. and ecology of fossil mammals as inferred from stable carbon [102] J. H. Burton, T. D. Price, L. Cahue, and L. E. Wright, “)e use and nitrogen isotope biogeochemistry: implications for of barium and strontium abundances in human skeletal Pleistocene bears,” Palaeogeography, Palaeoclimatology, tissues to determine their geographic origin,” International Palaeoecology, vol. 107, no. 3-4, pp. 213–225, 1994. Journal of Osteoarchaeology, vol. 13, no. 1-2, pp. 88–95, 2003. [118] H. Bocherens, D. Billiou, M. Patou-Mathis, D. Bonjean, [103] D. Mueller and E. Heinzle, “Stable isotope-assisted metab- M. Otte, and A. Mariotti, “Paleobiological implications of the olomics to detect metabolic flux changes in mammalian cell isotopic signatures (13C, 15N) of fossil mammal collagen in cultures,” Current Opinion in Biotechnology, vol. 24, no. 1, Scladina Cave (Sclayn, Belgium),” Quaternary Research, pp. 54–59, 2013. vol. 48, no. 3, pp. 370–380, 1997. [104] T. Tütken and T. Vennemann, “Stable isotope ecology of [119] H. Bocherens, M. Stiller, K. A. Hobson et al., “Niche Miocene large mammals from Sandelzhausen, southern partitioning between two sympatric genetically distinct Germany,” Paläontologische Zeitschrift, vol. 83, no. 1, cave bears (Ursus spelaeus and Ursus ingressus) and brown pp. 207–226, 2009. bear (Ursus arctos) from Austria: isotopic evidence from [105] L. A. Gregoricka, “Residential mobility and social identity in fossil bones,” Quaternary International, vol. 245, no. 2, the periphery: strontium isotope analysis of archaeological pp. 238–248, 2011. tooth enamel from southeastern Arabia,” Journal of Ar- [120] D. Fernandez-Mosquera,´ M. Vila-Taboada, and A. Grandal- chaeological Science, vol. 40, no. 1, pp. 452–464, 2013. d’Anglade, “Stable isotopes data (δ13C, δ15N) from the cave [106] E. McManus, J. Montgomery, J. Evans, A. Lamb, R. Brettell, bear (Ursus spelaeus): a new approach to its palae- and J. Jelsma, “To the land or to the sea: diet and mobility in oenvironment and dormancy,” Proceedings of the Royal Early Medieval Frisia,” Journal of Island and Coastal Ar- Society of London B: Biological Sciences, vol. 268, no. 1472, chaeology, vol. 8, no. 2, pp. 255–277, 2013. pp. 1159–1164, 2001. [107] J. E. Ericson, “Strontium isotope characterization in the [121] M. P. Richards, M. Pacher, M. Stiller et al., “Isotopic evidence study of prehistoric human ecology,” Journal of Human for omnivory among European cave bears: late Pleistocene Evolution, vol. 14, no. 5, pp. 503–514, 1985. Ursus spelaeus from the Pes¸teracu Oase, Romania,” Pro- [108] S. Hillson, Dental Anthropology, Cambridge University ceedings of the National Academy of Sciences, vol. 105, no. 2, Press, Cambridge, UK, 1996. pp. 600–604, 2008. 20 Journal of Analytical Methods in Chemistry

[122] Z. Nerudova,´ M. Nývltova´ Fiˇsakov´ a,´ and J. M´ıkova,´ “Palaeoenvironmental analyses of animal remains from the K˚ulnaCave (Moravian Karst, Czech Republic),” Quartar¨ , vol. 61, 2014. [123] O. Loreille, L. Orlando, M. Patou-Mathis, M. Philippe, P. Taberlet, and C. Hänni,“Ancient DNA analysis reveals divergence of the cave bear, Ursus spelaeus, and brown bear, Ursus arctos, lineages,” Current Biology, vol. 11, no. 3, pp. 200–203, 2001. [124] L. Orlando, D. Bonjean, H. Bocherens et al., “Ancient DNA and the population genetics of cave bears (Ursus spelaeus) through space and time,” Molecular Biology and Evolution, vol. 19, no. 11, pp. 1920–1933, 2002. [125] M. Pacher and A. J. Stuart, “Extinction chronology and palaeobiology of the cave bear (Ursus spelaeus),” Boreas, vol. 38, no. 2, pp. 189–206, 2009. [126] A. Berton, M. Bonato, A. Borsato et al., “Nuove datazioni radiometriche con il metodo U/) sulle formazioni sta- lagmitiche di Grotta all’Onda,” Rivista di Scienze Preistoriche, vol. 53, pp. 241–256, 2003. [127] G. Molara, “Resti faunistici provenienti dai livelli del Pleisto- cene superior di Grotta all’Onda (Camaiore, Lucca),” in Atti del 6° Convegno Nazionale di Archeozoologia, pp. 57–62, Centro visitatori del Parco dell’Orecchiella, Lucca, Italy, 2009. [128] T. Adani, Applicazioni delle sistematiche geochimiche ed isotopiche a reperti fossili Quaternari, M.S. thesis, Earth Sciences Department, Universitàdegli Studi di Firenze, 2013. [129] G. Cortecci and L. Lupi, “Carbon, oxygen and strontium isotope geochemistry of carbonates rocks from the Tuscan Nappe, Italy,” Mineral Petrol Acta, vol. 37, pp. 63–80, 1994. Hindawi Journal of Analytical Methods in Chemistry Volume 2018, Article ID 8274131, 8 pages https://doi.org/10.1155/2018/8274131

Research Article Development and Validation of an LC-MS/MS Method and Comparison with a GC-MS Method to Measure Phenytoin in Human Brain Dialysate, Blood, and Saliva

Raphael Ho¨sli ,1,2 Stefan Ko¨nig,3 and Stefan F. Mu¨hlebach 1

1Clinical Pharmacy and Epidemiology, Hospital Pharmacy, University of Basel, Spitalstrasse 26, CH-4031 Basel, Switzerland 2Spitalzentrum Biel, Apotheke, Vogelsang 84, CH-2501 Biel-Bienne, Switzerland 3Division of Forensic Medicine, University of Bern, B¨uhlstrasse 20, CH-3012 Bern, Switzerland

Correspondence should be addressed to Stefan F. M¨uhlebach; [email protected]

Received 12 December 2017; Revised 7 February 2018; Accepted 25 February 2018; Published 1 April 2018

Academic Editor: Federica Bianchi

Copyright © 2018 Raphael Hösli et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Phenytoin (PHT) is one of the most often used critical dose drugs, where insufficient or excessive dosing can have severe consequences such as seizures or toxicity. +us, the monitoring and precise measuring of PHT concentrations in patients is crucial. +is study develops and validates an LC-MS/MS method for the measurement of phenytoin concentrations in different body compartments (i.e., human brain dialysate, blood, and saliva) and compares it with a formerly developed GC-MS method that measures PHT in the same biological matrices. +e two methods are evaluated and compared based on their analytical performance, appropriateness to analyze human biological samples, including corresponding extraction and cleanup procedures, and their validation according to ISO 17025/FDA Guidance for Industry. +e LC-MS/MS method showed a higher performance compared with the GC-MS method. +e LC-MS/MS was more sensitive, needed a smaller sample volume (25 µL) and less chemicals, was less time consuming (cleaning up, sample preparation, and analysis), and resulted in a better LOD (<1 ng/mL)/LOQ (10 ng/mL). +e calibration curve of the LC-MS/MS method (10–2000 ng/mL) showed linearity over a larger range with correlation coefficients r2 > 0.995 for all tested matrices (blood, saliva, and dialysate). For larger sample numbers as in pharmacokinetic/pharmacodynamic studies and for bedside as well as routine analyses, the LC-MS/MS method offers significant advantages over the GC-MS method.

1. Introduction Phenytoin (PHT) belongs to the most widely prescribed drugs to prevent and control most types of seizure disorders Sensitive and specific quantification methods are of critical and to treat epilepsy [3]. It is one of the most often used importance when monitoring individualized drug therapy in critical dose drugs where insufficient or excessive dosing can patients or investigating drug concentration in forensic have severe consequences such as seizures or toxicity. +us, toxicology [1]. Critical dose drugs but also newly developed the monitoring and precise measuring of PHT concentra- and designed complex drugs require analytical methods to tions in patients is crucial [4, 5]. As an example, in forensic check for effective drug delivery to target tissues and to toxicology, epilepsy patients under PHT treatment who have minimize toxicity in sensitive organs or cells. When such been involved in an accident have to be analyzed in order to drugs have to be used in patients with varying pharmaco- verify whether the PHT concentration was adequate or kinetics (PK) (e.g., ICU patients), an appropriate therapeutic possibly the reason for the accident [6]. However, there are drug monitoring (TDM), which allows, for example, to several characteristics of PHT including a relatively low correlate the drug concentration in easy accessible plasma therapeutic index, difficult pharmacokinetics (PK) and phar- samples with those in the tissue of action, becomes even macodynamics (PD), saturable oxidative biotransformation, more relevant for a safe and efficient drug treatment [2]. and the nonlinear clearance, which complicate a therapeutic 2 Journal of Analytical Methods in Chemistry drug monitoring (TDM) aimed at preventing intoxication of Artificial cerebrospinal fluid (aCSF; dialysate solution) was patients or treatment failures [7]. prepared according to M Dialysis AB (Stockholm, Sweden) +us, researchers and practitioners are interested in [14]. Blood CPDA-1 (anticoagulant citrate phosphate dex- specific, sensitive, robust, and cost-effective methods to trose adenine solution; to simplify only named blood in the identify PHT concentrations in patients. +ereby, several following) was obtained from the Blood Donor Center compartments to measure the PHT concentration could be (Bern, Switzerland). Saliva was obtained from one of the addressed such as blood, saliva, and CNS fluid (micro- investigators. 20–60 μL PHT-containing samples from pa- dialysate). +e correlation of PHT in different body com- tients collected from a 2 µL/min flow rate brain microdialysis partments is not yet completely understood and has only and 2 mL of CPDA containing PHT patient blood samples recently been addressed by researchers who have compared were provided by the Department of Neurosurgery (Kant- the measurement of PHT in these different compartments onsspital Aarau AG, Switzerland and Centre Hospitalier with a GC-MS method [8]. While the GC-MS has long been Universitaire Vaudois, Switzerland). All biological samples the standard method in forensic testing, LC-MS/MS methods (blood and dialysates) were frozen and stored at −24°C. have become more common, as they generally offer some Before sample analysis, the samples were thawed at room advantages over GC-MS [9]. Recently, researchers have de- temperature for 30 minutes and homogenized by shaking veloped an LC-MS/MS method to measure PHT in one with a vortex for one minute. specific body compartment (i.e., blood plasma or serum) [10]. Missing, however, is a thorough comparison of the perfor- 2.2. Internal Standards, Calibrator Standard System Suit- mance of these two analytical methods in the detection and ability Testing, and Sample Preparation. +e internal stan- analysis of PHT in different body compartments (i.e., blood, dard (IS) stock solution was prepared by adding 100 µL of saliva, and samples from brain tissue microdialysis). the PHT-D (100 µg/mL) to 9900 µL of MeOH. 5 mL of this +e aim of the present study was to develop and validate 10 solution was added to 95 mL of 1 M perchloric acid aqueous an LC-MS/MS method for the measurement of PHT con- solution to get the final concentration of 50.0 ng/mL, which centrations in different body compartments such as blood is used as IS working solution. +e PHT reference stock and saliva, as well as samples from brain tissue microdialysis solution (1.00 mg/mL) was used to obtain eight calibration often used in neurology and neurosurgery, where antiepi- (Cal) solutions with concentrations of 2000, 1000, 500, 250, leptic therapy is often mandatory [11, 12], and to compare its 100, 50, 20, and 10 ng/mL PHT. 20 µL of these Cal solutions efficiency with a formerly developed GC-MS method [8]. were added to 980 µL of the biomatrices to get the Cal +e fact that this established GC-MS method measured PHT working solutions. For quality control (QC), solutions with in the same biological matrices (i.e., blood, saliva, and hu- 1600, 400, 30, and 10 ng/mL PHT were prepared out of PHT man brain dialysate) enables a reliable comparison with reference stock solution (1.00 mg/mL). regard to the performance of GC-MS versus LC-MS/MS in +e IS working solution of 75 µL was added either to an measuring PHT in different body compartments. +e two aliquot of 25 µL Cal working solution, QC solutions, or 25 µL methods are evaluated and compared based on their ana- sample from patients containing PHT. +e sample prepa- lytical performance, appropriateness to analyze human bi- ration for the LC-MS/MS consisted of pipetting 75 µL of IS ological samples, including corresponding extraction and working solution to 25 µL sample into a deep well plate cleanup procedures, and their validation according to ISO (0.6 mL, Chemie Brunschwig AG, Basel, Switzerland) cov- 17025/FDA Guidance for Industry [13]. Finally, the suit- ered by a sealing mat (Silicone, Chemie Brunschwig AG, ability of the two analytical methods for PK/PD studies, Basel, Switzerland). +e well plates were rigorously shaken bedside measurement, and forensic use is discussed. In for 5 minutes and then centrifuged (4.500 U/min; Mikro addition, the LC-MS/MS method developed in the current 22R, Hettich Instruments, Andreas Hettich AG, Bach,¨ study is compared with an established LC-MS/MS method Switzerland) for 30 minutes at about 8°C (Figure 1). +e which measured PHT in blood plasma samples [10]. processed samples were ready for the LC-MS/MS analysis. 2. Materials and Method 2.3. LC-MS/MS Settings. +e prepared samples were placed 2.1. Chemicals and Samples Used for the Development of the into the autosampler (Dionex WPS-3000TSL Olten, Swit- LC-MS/MS Method and Its Validation. PHT reference zerland) which was set at 8°C. With a100 µL syringe from the substance was purchased from Desitin Pharma GmbH autosampler, 10 µL of the prepared samples was injected into (Liestal, Switzerland) and from the European Pharma- a 130 µL loop. +e solvent rack (Dionex SRD-3600, Olten, copoeia (PHT Ph. Eur. Standard, EDQM, Strasbourg, Switzerland) carried the mobile phase A (H2O + HCOOH France). +e IS for LC-MS/MS was PHT-D10 (PHT D-10, (100 + 0.1, v + v)) and phase B (MeCN + HCOOH (100 + 0.1, v C15H2D10N2O2, MW � 262.33) in methanol (MeOH) + v)). +ese mobile phases were delivered by three pumps (100 μg/mL) from Cerilliant (Round Rock, TX). (binary pump 1 (flow 0.350 mL/min) and isocratic pump 2 Calcium chloride, perchloric acid, citric acid mono- (flow 0.200 to 1.000 mL/min) (Dionex pump HPG-3200A, hydrate, potassium chloride, magnesium chloride hexahy- Olten, Switzerland), and binary pump 3 (Dionex pump ISO- drate, sodium chloride, sodium hydroxide, and the solvents 3100A, Olten, Switzerland)) connected to a triple stage (methanol, acetic acid 100%, and acetone) were of analytical quadrupole mass spectrometer with linear ion-trap capability grade and purchased from Merck (Darmstadt, Germany). (3200 QTrap, Analyst Software Version 1.5.1, Applied Journal of Analytical Methods in Chemistry 3

25 μL sample and 75 μL of HCIO with 4 e DWP was D -PHT was 10 certrifuged for 30 min pipetted into the (4500 U/min) at 8ºC deep well plate (DWP)

e centrifuged DWP was placed into the LC- MS autosampler and 10 μL of the worked-up sample was injected

Figure 1: Sample preparation for the LC-MS/MS analyses for blood, saliva, and aCSF samples.

Table 1: Settings of the HPLC program. Pump 1 (main column (MC)) Pumps 2 and 3 (trapping column (TC)) Time Flow Flow Flow pump 5 (minutes) %B Comments %B Switching valve (µL/min) (µL/min) (H2O + 0.1% HCOOH) (µL/min) 0 35 Start MS and pumps 50 300 800 TC → waste, MC → MS (loading) 0.5 35 Start gradient 50 300 800 TC → MC → MS (eluting) 0.6 ↓ 500 1 97.5 50 20 20 2 97.5 TC → waste, MC → MS 2.5 35 50 300 800 Reequilibration

Biosystems/MDS Sciex, Toronto, Canada) (Table 1). For the including selectivity, sensitivity, accuracy, recovery of PHT, mass spectrometric detection, SRM scan mode (selective reproducibility and suitability of the calibration curves, reaction monitoring) was used. SRM transitions and mass stability of PHT, and matrix effects. +e selectivity and spectrometric conditions were as follows: transition: sensitivity (absence of PHT) were verified by analyzing blank 253.1 → 182.2 (PHT) and 253.1 → 192.2 (PHT-D10); orifice (V): samples without PHT (extraction and matrix effects). For the 36; collision energy (eV): 41 (PHT) and 51 (PHT-D10); and accuracy, QCs and Cal samples were analyzed. +e recovery dwell time (msec): 100. Electrospray ionization was per- of PHT was analyzed by measuring QCs at different levels. formed in positive ion mode for the analyte and the IS. +e +e reproducibility and suitability of the calibration curves following instrument parameters for ionization were used: ion was measured by a complete series of Cal 1 to Cal 8 source voltage: 5000 volt, curtain gas: 25, gas 1: 40 and gas (LC-MS/MS) analyses. +e limit of detection (LOD) and 2: 60; and the CAD gas was set to 5 (arbitrary units for the the limit of quantification (LOQ) were analyzed using Cal 1 gas settings). As trapping column, a Phenomenex Gemini (LC-MS/MS 10 ng/mL PHT). +e LOD was checked as a Polar column (2.0 ×10 mm, 5 µm; Brechbu¨hler AG, Schlieren, signal-to-noise (S/N) ratio of more than 4 :1. +e LOQ was Switzerland) temperated to room temperature was used. +e considered as 5 times the response to a blank sample. +e main column Phenomenex Synergy Polar RP column stability tests consisted of the freeze-thaw stability of PHT, (2.0 × 50 mm; Brechbu¨hler AG, Schlieren, Switzerland) was which was determined after 3 freeze-thaw cycles. +e short- placed into the column oven (Cluzeau Info Labo CrocoCil) term stability was analyzed by keeping the samples thawed at set on 50°C with a column thermostat (Dionex TCC-3100, ambient temperature for at least 6 hours, frozen for at least Olten, Switzerland) including switching valve (Figure 2). 12 hours at −25°C ± 5°C, and again thawed, worked-up, and +is system was operated by Analyst Software (Version 1.5.1, analyzed. Postpreparative stability was evaluated to determine AB Sciex, Toronto, Canada). whether an analytical run can be reinjected in the case of instrument failure and, furthermore, whether the preparation of a large number of samples could be done at once. 2.4. Validation of the LC-MS/MS Method according to ISO +erefore, one of the validation runs was analyzed a second 17025/FDA Guidance for Industry. +e validation was car- time after 7 days. +e described criteria for Cal curves, QC, ried out according to ISO 17025/FDA Guidance for Industry accuracy, and precision had to be met. 4 Journal of Analytical Methods in Chemistry

Loading Isocratic pump

Analytical column T-union

ESI-MS/MS Autosampler column Trapping Waste

Binary pump 1 Binary pump 2

(a)

Eluting Analytical column Isocratic pump Analytical column T-union

ESI-MS/MS Autosampler column Trapping Waste

Binary pump 1 Binary pump 2

(b)

Figure 2: LC-MS/MS settings.

Matrix effects were analyzed by comparing the cali- min–max deviations of 1–8% for Cal 2 (20 ng/mL) to Cal 8 bration curves generated with the three matrices aCSF, (2000 ng/mL) with 3% for Cal 1 (10 ng/mL). +e calibration blood, and saliva. PHT microdialysis and blood samples curves for all three matrices were linear. +e regression co- 2 2 from patients were analyzed to demonstrate the suitability of efficients (r ) of the three different matrices were r blood � 0.996 2 2 the method for biological samples from patients. (n � 3), r dialysate � 0.997 (n � 6), and r saliva � 0.995 (n � 3). Reinjection after 7 days showed no difference in accuracy. +e sample volume needed was 25 µL. +e sample prep- 2.5. Comparison of the LC-MS/MS and the GC-MS aration time was about 2 min per sample (6 hours for Method. +e LC-MS/MS method was evaluated and com- 182 samples). +e run time for one LC-MS/MS analyses ¨ pared with the GC-MS method published by Hosli et al. [8] was 7 min. with regard to its analytical performance, appropriateness to analyze human biological samples, including corresponding extraction and cleanup procedures, and its validation 3.2. Comparison of the LC-MS/MS with the GC-MS according to ISO 17025/FDA Guidance for Industry. Method. After validation of the LC-MS/MS method, it +e statistical data were calculated with Microsoft Excel was compared with the referred GC-MS method [8]. Table 2 and IBM SPSS Statistics 22. To compare the different ma- shows the comparative results of the two methods for their trices, a one-way ANOVA was calculated. +e corre- analytical performance, appropriateness to analyze human sponding values were checked for significance by t-tests. biological samples, including corresponding extraction and cleanup procedures, and its validation according to ISO 3. Results 17025/FDA Guidance for Industry (Table 2). +e selectivity and the sensitivity were met by both 3.1. Validation of the LC-MS/MS Method. +e retention time methods, and the recovery showed no differences (Table 2). (RT) for PHT and for PHT-D10 (IS) was about 2.8 min But the accuracy differed between the two methods. +e GC- (Figure 3). +e selectivity and sensitivity were checked; all MS method showed a higher variation at Cal 1 (20%) than blank samples were negative. +e recovery of PHT after the LC-MS/MS method (Cal 1 � 3%). As expected, the precipitation with HClO4 was 89.5% for QC1 (10 ng/mL) biggest difference in terms of analytical performance and 97.1% for QC3 (1600 ng/mL) compared to the amount between the two methods was observed by the LOQ (GC- found in unprepared samples (�100%). +e LOD calculated MS � 50 ng/mL; LC-MS/MS � 10 ng/mL) and LOD (GC- as S/N ratio of 4 :1 for this method in aCSF, saliva, and blood MS � 15 ng/mL; LC-MS/MS � <1 ng/mL) (Table 2). was set at <1 ng/mL. +e LOQ calculated as 5 times the Both methods showed linear regression coefficients (r2) response/blank was 10 ng/mL PHT. For the accuracy, the higher than 0.995 in all three different matrices for the PHT Cal 1 to Cal 8 were assessed. +e calibrator values showed calibration curve. +e calibration range of the LC-MS/MS Journal of Analytical Methods in Chemistry 5

Max. 2010.0 cps Max. 2.8e5 cps

2000 2.86 2.8e5 2.86 2.5e5 1500 2.0e5 1.5e5 1000 LOQ QC3 1.0e5 Intensity (cps) Intensity Intensity (cps) Intensity 500 5.0e4

0 0.0 1.0 2.0 3.0 4.0 5.0 1.0 2.0 3.0 4.0 5.0 Time (min) Time (min) XIC of +MRM (2 pairs): 253.103/182.200 Da... XIC of +MRM (2 pairs): 253.103/182.200 Da... Max. 5470.0 cps Max. 5320.0 cps

5470 2.86 2.85 5000 5000

4000 4000

3000 IS 3000 IS

2000 2000 Intensity (cps) Intensity Intensity (cps) Intensity 1000 1000

0 0 1.0 2.0 3.0 4.0 5.0 1.0 2.0 3.0 4.0 5.0 Time (min) Time (min) XIC of +MRM (2 pairs): 263.103/192.200 Da... XIC of +MRM (2 pairs): 263.103/192.200 Da... Figure 3: Chromatogram of phenytoin (illustrated for LOQ (10 ng/mL) and QC3 (400 ng/mL)) with PHT-D10 as IS (50 ng/mL).

Table 2: Comparison of the GC-MS [8] versus LC-MS/MS method. Criterion GC-MS LC-MS/MS

Retention time PHT 15.12 min, IS MPPH 16.15 min PHT and PHT-D10 2.8 min Good peak differentiation and quantification Selectivity/sensitivity All blank samples were negative of PHT. All blank samples were negative (absence of PHT) (no presence of PHT) (no presence of PHT) 94.1% for QC2 (100 ng/mL) 89.5% for QC1 (10 ng/mL) Recovery 94.3% for QC5 (1000 ng/mL) 97.1% for QC3 (1600 ng/mL) LOD (calculated as S/N ratio of 4 :1) 15 ng/mL <1 ng/mL LOQ (calculated as 5 times the 50 ng/mL 10 ng/mL response/blank) +e calibrator values showed min–max +e calibrator values showed min–max Accuracy percent deviations of 1–20% for percent deviations of 1–8% for Cal 1 (50 ng/mL) to Cal 6 (1200 ng/mL) Cal 1 (10 ng/mL) to Cal 8 (2000 ng/mL) 2 2 r blood � 0.998 (n � 2) r blood � 0. 996 (n � 3) 2 2 2 Regression coefficient, r r dialysate � 0.999 (n � 8) r dialysate � 0.997 (n � 6) 2 2 r saliva � 0.999 (n � 2) r saliva � 0.995 (n � 3) Calibration range 50–1200 ng/mL 10–2000 ng/mL Run time per analysis 30 min 7 min Injection volume of the sample 2.0 µL 10 µL Sample preparation time 5 h for 25 samples 6 h for 182 samples Dried extracts were stable for ≥4 weeks Reinjection after 7 days showed Stability of the processed samples (min/max deviation 4%). No effect by reinjection no difference in accuracy and storage (33 h) on the autosampler Sample volume needed 50 µL 25 µL

(from 10 ng/mL to 2000 ng/mL) is twice as large as of the for both methods. +e sample preparation procedure is GC-MS (50 ng/mL to 1200 ng/mL). +e stability of the demonstrated in Figure 1 for LC-MS/MS and Figure 4 for samples after extraction and cleaning up was demonstrated GC-MS. 6 Journal of Analytical Methods in Chemistry

Outlet (dialysate) SPE (1) Acetonitrile (1 ml) (2) Citric buﬀer pH 5 (1 ml) aCSF (3) Sample (4) Citric buﬀer pH 5 (1 ml) Aliquot of (5) Acetic acid (1 ml) 50 µl+ (6) Elute with acetone (2x with 1 ml) 450 µl Inlet matrix (perfusate) Sample +IS Flow rate collected (2 µl/min) during 30 min Reconstitution and methylation (derivatization)

Semipermeable membrane Injection Saliva of 2 µl sample Blood sample

Figure 4: Sample preparation for the GC-MS analyses [8].

3.3. Comparison of the LC-MS/MS Method with a Formerly Concerning the LOD, there was a huge difference be- Established LC-MS/MS Method. Recently, a LC-MS/MS tween the two methods. +e LOD of the LC-MS/MS method method has been developed which measures PHT in blood was 15 times better than the one of the GC-MS methods: the plasma or serum [10]. For the measurement of PHT in blood, LOD of the LC-MS/MS method was <1 ng/mL compared to the newly validated LC-MS/MS method can hence also be 15 ng/mL for the GC-MS method (increments by a factor of compared with this recently published study. +e two ten). Similarly, the difference in LOQ was 5 times lower in methods show some similarities such as an identical IS LC-MS/MS (10 ng/ml) compared to GC-MS (50 ng/mL). +e (100 µg/mL PHT-d10), similar sample volumes needed (25 µL LOQ for the LC-MS/MS could be set even lower than versus 20 µL [10]), and a comparable retention time (2.8 min 10 ng/ml PHT (Cal 1). +e FDA guidelines which claim versus approximately 2.1 min [10]). Both methods showed a minimal reproducibility at the LOQ level of 20% were well linear regression coefficients (r2) higher than 0.99 in the blood below (deviation to target PHT amount: <8% in aCSF (n � 6), matrix. +e accuracy was similar as both studies showed <4% in blood (n � 3), <9% in saliva (n � 3); accuracy: aCSF deviations of <10%. With regard to the calibration range and 103%, blood 101%, and saliva 106%). +e LOQ of the GC- the calibration solution, the two LC-MS/MS methods differ. MS method and hence the lowest concentration level (Cal 1 While the LC-MS/MS method developed in this study showed at 50 ng/ml) of the calibration curve showed a deviation a calibration range from 10 ng/mL to 2000 ng/mL, the cali- value of 19%. +e LC-MS/MS method, in contrast, showed bration curve of the published LC-MS/MS method [10] a value of only 3% deviation at the lowest Cal (10 ng/mL). ranged from 100 ng/mL to 4000 ng/mL. +e calibration so- +is difference is of high importance, as samples with even lution in the current study was the respective biological matrix lower concentrations could be reliably analyzed. (e.g., blood). In the published study [10], phosphate-buffered +e calibration range (from 10 ng/mL to 2000 ng/mL) of saline was used as the calibration solution. the LC-MS/MS method was twice as large as of the GC-MS method (50 ng/mL to 1200 ng/mL). +is indicates that the 4. Discussion LC-MS/MS method is more powerful and effective over a larger range of concentration, since the linearity is given In this study, a LC-MS/MS method to measure PHT in over a larger area (10 ng/mL–2000 ng/mL) compared to the different biological samples was successfully validated and GC-MS method (50 ng/mL–1200 ng/mL). compared with a similarly validated GC-MS method [8]. As IS, two different substances were used. MPPH as Overall, the LC-MS/MS method showed to be a more a structurally related compound was used for the GC-MS specific analytical method with a higher general perfor- method. As IS for LC-MS/MS, deuterated PHT (PHT-D10) mance (Table 2). +e LC-MS/MS method needed less sample was used, which is the same molecule as PHT and differs only volume, less chemicals, and less analytical time and therefore by the molecular mass (+1). All the physicochemical pro- resulted in less costs for the sample preparation. cesses upon cleanup and analysis are identical or highly Journal of Analytical Methods in Chemistry 7

similar for PHT and PHT-D10. MMPH, however, could be Finally, the appropriateness of the method also depends on chemically affected in a different way than PHT, which could the biological matrix. Both methods can generally be used to lead to a systematic bias in a given situation [15]. measure PHT in blood and saliva, as the sample volume is less Regarding the sample preparation procedure, the LC- limiting. As mentioned before, however, for dialysates, the MS/MS (Figure 1) showed an important advantage com- most difficult aspect is to get enough sample volume. +erefore, pared to the GC-MS method as it only needs 3 steps of the LC-MS/MS method needing only half of the sample volume sample preparation compared to 11 steps necessary for the compared to the GC-MS method is more suited for micro- GC-MS method including a solid-phase extraction (SPE) dialysate measurements. With respect to the LOD/LOQ, the and derivatization with a more critical chemical trime- LC-MS/MS method is also better suited for PK/PD studies, as it thylsulfonium (TMSH) (Figure 4). +is resulted in signifi- allows to include patients with low PHT dosages. cant shortening of the overall analysis: Preparation of the In addition, the newly established LC-MS/MS method samples before injection for GC-MS is about ten times more was compared with a recently published LC-MS/MS method time consuming than for the LC-MS/MS. For the GC-MS [10]. While this study measured PHT only in one body method, researchers needed 5 hours to prepare 25 samples (5 compartment (i.e., blood plasma or serum), the current LC- samples/h), whereas for the LC-MS/MS method 182 samples MS/MS method was developed and validated for the mea- were prepared in 6 hours (30.3 samples/h), which corre- surement of PHT in different body compartments (i.e., blood, sponds to 6 times the amount of prepared samples per hour saliva, and samples from brain tissue microdialysis often used compared to the GC-MS method. in neurology and neurosurgery). +e calibration range of the From the exposure side, the volumes are much larger published LC-MS/MS method [10] (from 100 ng/mL to and the exposure to the chemicals are more prolonged with the 4000 ng/mL) is appropriate for the measurement of PHT in GC-MS method compared to the sample cleanup for the LC- blood plasma. As the PHT concentrations in brain tissue MS/MS method. Especially, the derivatization agent TMSH is dialysates are much smaller than in blood plasma, the LC- critical to handle because of toxicity. +e risk of serious and MS/MS method of the current study was more appropriate for even irreversible effects through inhalation, skin contact, or such samples, showing a lower calibration range from eye exposure is well known. TMSH is also considered to be 10 ng/mL to 2000 ng/mL. Finally, as the aim of this study was teratogenic. +erefore, the potential health risk for the labo- to measure PHT in different biological matrices, a general ratory staff handling the samples can be reduced by the LC- substitute solution for blood plasma such as phosphate- MS/MS method and the elimination of a safety critical agent. buffered saline [10] could not be used. Instead, the fluid of +e amount of biological samples needed for the GC-MS the respective body compartment was used as calibration method (50 µL) was twice as much as for the LC-MS/MS solution (e.g., artificial cerebrospinal fluid (aCSF) for the (25 µL). +e sample volume is a critical point for PK/PD measurement of PHT in the brain tissue dialysates). +is also studies, where, for example, by continuing dialysis from eliminates a potential analytical bias due to matrix effects. brain in neurosurgical patients only small volumes of samples per time point/period are available. For 50 µL di- 5. Conclusion alysate about 25 minutes collecting time is necessary at the usual flow rate of ∼2 µL per minute [12, 16]. +erefore, not In this study, a LC-MS/MS method to measure PHT in a requested specific time point, but a rather large time different biological samples (i.e., human brain dialysate, segment is represented which can influence the requested blood, and saliva) was developed and validated under cir- results. +e reduced sample volume needed (25 µL) for the cumstances that ensured a high comparability with an LC-MS/MS analyses reduces the dialyses time needed per established GC-MS method [8]. Overall, the study concludes sample to about 15 minutes. +e smaller the dialysis time, that LC-MS/MS is not only better performing in human the more precise correlations of the respective tissue con- PHT concentration measuring or comparable drug PK/PD centration with plasma/blood samples can be made. studies but is the only one to be used for bedside analysis. Furthermore, LC-MS/MS also has the shorter run time. +e time-consuming sample preparation and the long run +e time needed for 100 GC-MS analyses would be ap- time of the GC-MS method delay the result, which is critical proximately 50 hours. +e LC-MS/MS method, in contrast, in TDM. +e higher sensitivity, the smaller needed sample needs only 11 hours and 40 minutes for 100 analyses. +is is volume, the better LOD/LOQ, the less time-consuming a time saving of more than 38 hours. While this may not be cleaningup and sample preparation procedure, and the highly relevant for forensic purposes, for bedside and shorter run time make the LC-MS/MS method the preferred routine analyses (real-time) and PK/PD studies with larger analytical procedure. numbers of samples, this factor is relevant. Also, when the time between taking a sample and the result needed is short, Conflicts of Interest as it is in TDM to adjust subsequent dosing for PHT +e authors declare that they have no conflicts of interest. treatment, this time saving is crucial. +e costs for one way materials per sample was about 50% lower for the LC-MS/MS compared to the GC-MS References method. Especially because no SPE device was needed. Also, [1] S. Deeb, D. A. McKeown, H. J. Torrance, F. M. Wylie, the reduced work load for the laboratory technician must be B. K. Logan, and K. S. Scott, “Simultaneous analysis of 22 considered as an imported cost factor. antiepileptic drugs in postmortem blood, serum and plasma 8 Journal of Analytical Methods in Chemistry

using LC–MS-MS with a focus on their role in forensic cases,” Journal of Analytical Toxicology, vol. 38, no. 8, pp. 485–494, 2014. [2] S. L. von Winckelmann, I. Spriet, and L. Willems, “+era- peutic drug monitoring of phenytoin in critically ill patients,” Pharmacotherapy, vol. 28, no. 11, pp. 1391–1400, 2008. [3] E. H. Grover, Y. Nazzal, and L. J. Hirsch, “Treatment of convulsive status epilepticus,” Current Treatment Options in Neurology, vol. 18, no. 3, 2016. [4] M. F. Wu and W. H. Lim, “Phenytoin: a guide to therapeutic drug monitoring,” Proceedings of Singapore Healthcare, vol. 22, no. 3, pp. 198–202, 2013. [5] A. Tobler, R. Hosli,¨ S. Mühlebach, and A. Huber, “Free phenytoin assessment in patients: measured versus calculated blood serum levels,” International Journal of Clinical Phar- macy, vol. 38, no. 2, pp. 303–309, 2016. [6] L. Nilsson, B. Y. Farahmand, P. G. Persson, I. +iblin, and T. Tomson, “Risk factors for sudden unexpected death in epilepsy: a case-control study,” =e Lancet, vol. 353, no. 9156, pp. 888–893, 1999. [7] E. Martin, T. N. Tozer, L. B. Sheiner, and S. Riegelman, “+e clinical pharmacokinetics of phenytoin,” Journal of Phar- macokinetics and Biopharmaceutics, vol. 5, no. 6, pp. 579–596, 1977. [8] R. Hösli,A. Tobler, S. König,and S. Mühlebach,“A quantitative phenytoin GC-MS method and its validation for samples from human ex situ brain microdialysis, blood and saliva using solid-phase extraction,” Journal of Analytical Toxicology, vol. 37, no. 2, pp. 102–109, 2013. [9] E. R. Perez, J. A. Knapp, C. K. Horn, S. L. Stillman, J. E. Evans, and D. P. Arfsten, “Comparison of LC–MS-MS and GC–MS analysis of benzodiazepine compounds included in the drug demand reduction urinalysis program,” Journal of Analytical Toxicology, vol. 40, no. 3, pp. 201–207, 2016. [10] J. Peat, C. Frazee, and U. Garg, “Quantification of free phenytoin by liquid chromatography tandem mass spectrometry (LC/MS/MS),” Methods in Molecular Biology, vol. 1383, pp. 241–246, 2016. [11] E. P. +elin, L. H. Keri, P. J. Carpenter, and A. H. Hutchinson, “Microdialysis monitoring in clinical traumatic brain injury and its role in neuroprotective drug development,” AAPS Journal, vol. 19, no. 2, 2017. [12] Y. Yamamoto, M. Danhof, and C. M. Elizabeth de Lange, “Microdialysis: the key to physiologically based model pre- diction of human CNS target site concentrations,” AAPS Journal, vol. 19, no. 4, 2017. [13] FDA, Guidance for Industry, Bioanalytical Method Validation, U.S Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Re- search (CDER), Center for Veterinary Medicine (CVM), Rockville, MD, 2001. [14] Perfusion Fluid, 2017, http://www.mdialysis.com/iscus/iscus- international/products/catheter-accessories/perfusion-fluid. [15] J. Wieling, “LC-MS-MS experiences with internal standards,” Chromatographia Supplement, vol. 55, pp. 107–113, 2002. [16] H. K. Kimelberg, “Water homeostasis in the brain: basic concepts,” Neuroscience, vol. 129, no. 4, pp. 851–860, 2004. Hindawi Journal of Analytical Methods in Chemistry Volume 2018, Article ID 8341630, 9 pages https://doi.org/10.1155/2018/8341630

Research Article Analysis of Polycyclic Aromatic Hydrocarbons in Ambient Aerosols by Using One-Dimensional and Comprehensive Two-Dimensional Gas Chromatography Combined with Mass Spectrometric Method: A Comparative Study

Yun Gyong Ahn ,1 So Hyeon Jeon,1 Hyung Bae Lim,2 Na Rae Choi,3 Geum-Sook Hwang,1 Yong Pyo Kim,4 and Ji Yi Lee 3

1Western Seoul Center, Korea Basic Science Institute, Seoul 03759, Republic of Korea 2Air Quality Research Division, National Institute of Environmental Research, Incheon 22689, Republic of Korea 3Department of Environmental Science and Engineering, Ewha Womans University, Seoul 03759, Republic of Korea 4Department of Chemical Engineering and Material Science, Ewha Womans University, Seoul 03760, Republic of Korea

Correspondence should be addressed to Ji Yi Lee; [email protected]

Received 14 December 2017; Revised 6 February 2018; Accepted 19 February 2018; Published 1 April 2018

Academic Editor: Federica Bianchi

Copyright © 2018 Yun Gyong Ahn et al. /is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Advanced separation technology paired with mass spectrometry is an ideal method for the analysis of atmospheric samples having complex chemical compositions. Due to the huge variety of both natural and anthropogenic sources of organic compounds, simultaneous quantification and identification of organic compounds in aerosol samples represents a demanding analytical challenge. In this regard, comprehensive two-dimensional gas chromatography with time-of-flight mass spectrometry (GC×GC- TOFMS) has become an effective analytical method. However, verification and validation approaches to quantify these analytes have not been critically evaluated. We compared the performance of gas chromatography with quadrupole mass spectrometry (GC-qMS) and GC×GC-TOFMS for quantitative analysis of eighteen target polycyclic aromatic hydrocarbons (PAHs). /e quantitative obtained results such as limits of detection (LODs), limits of quantification (LOQs), and recoveries of target PAHs were approximately equivalent based on both analytical methods. Furthermore, a larger number of analytes were consistently identified from the aerosol samples by GC×GC-TOFMS compared to GC-qMS. Our findings suggest that GC×GC-TOFMS would be widely applicable to the atmospheric and related sciences with simultaneous target and nontarget analysis in a single run.

1. Introduction policy [7]. /e EPA priority 16 PAHs and two additional PAHs are now being monitored by European agencies, and they have Human health research associated with polycyclic aromatic sought to quantify the individual concentrations of benzo[e] hydrocarbons (PAHs) has raised concerns because certain pyrene and perylene in environmental samples [6]. PAHs are PAHs are classified as probable human carcinogens [1–4] and found in ambient air in the gas phase and as sorbents to have shown tumorigenic activity and endocrine disrupting aerosols [8]. /us, air monitoring of PAHs to quantify in- activity in mammals [5]. /e US EPA has included 16 of them halation exposure and to identify other organic compounds is in the list of priority pollutants and has established a maximum important for insight into photochemical reactions. /e contaminant level of 0.2 μg/L for benzo[a]pyrene in drinking quantification and identification of organic compounds in air water [6]. In the European Union (EU), eight PAHs have been samples is an important feature of atmospheric chemistry and identified as priority hazardous substances in the field of water represents some demanding analytical challenges [9]. 2 Journal of Analytical Methods in Chemistry

For these reasons, a key issue in current analytical (>99%) were purchased from Aldrich (St. Louis, MI, USA), methods is the ability to measure a large number of com- and a standard mixture of eighteen PAHs was prepared at pounds with quantitative analysis for target analytes. a concentration of 1000 µg/mL. Deuterium-labeled internal Comprehensive two-dimensional gas chromatography standards of seven PAHs were purchased from Aldrich (GC×GC) coupled with mass spectrometry (MS) can screen (St. Louis, MI, USA) and Chiron (Trondheim, Norway) and for nontarget compounds with fast identification of the used for the spiking test as listed in Table 1. Working compounds in an entire sample [10]. /erefore, previous standard solutions (0.01∼10 µg/mL) were prepared and then studies applied GC×GC-MS for the identification of nu- stored at −20°C prior to use. merous compounds present in air samples [11–13]. How- ever, there are limitations on the validation of simultaneous quantification and identification of analytes in air samples. 2.3. Preparation of Samples. Air sampling filters were Correspondingly, a validation of simultaneous identification extracted with a mixture of dichloromethane and methanol and quantification of PAHs and other compounds in air (3 :1, v/v) two times using an accelerated solvent extractor samples by GC×GC–MS is required. A TOF mass spec- (ASE) (Dionex ASE-200) at 40°C and 1700 psi for 5 min. trometer was used to acquire sufficient data from a com- Prior to the extraction, seven deuterated internal standards prehensive two-dimensional chromatographic technique (Nap-d8, Ace-d10, Phen-d10, Fla-d10, Chr-d12, Per-d12, that generated multiple narrow peaks from the short sec- and BghiPer-d12) were spiked in the filters to compensate ondary column [14, 15]. Generally, GC coupled with for matrix effects during the extraction procedure. Extracts quadrupole MS (GC-qMS) in the selected ion monitoring were blown down to 1 mL using a nitrogen evaporator (SIM) mode has been used for quantitative analysis of PAHs (TurboVap II, Caliper Life Sciences). GC×GC-TOFMS in air samples because of its selective detection for specific analysis was carried out using an Agilent GC (Wilmington, target compounds [16, 17]. However, a GC×GC-TOFMS Delaware, USA)-Quad-jet thermal modulation Pegasus 4D validated method suitable for the quantification of target TOFMS (LECO, St. Joseph, MI, USA). /e sample was in- PAHs in an aerosol sample compared with GC-qMS in the jected in the splitless mode at 300°C. /e GC×GC columns SIM mode has not yet been reported. /e aim of this study were as follows: DB-5MS (30 m × 0.25 mm ID, film thickness was to evaluate the effectiveness of GC×GC-TOFMS in the of 0.25 μm) and 1.17 m DB-17MS (0.18 mm OD, 0.18 μm quantitative analysis of target PAHs as well as the fast film). /e operating conditions of GC-MS and GC×GC- identification of multiple compounds for aerosol samples. TOFMS are summarized in Table 2. /e validity of the quantitative results obtained by both GC×GC-TOFMS and GC-qMS in the SIM mode was 3. Results and Discussion demonstrated by several method performance parameters such as linearity, accuracy, and repeatability. 3.1. GC-qMS and GC×GC-TOFMS for Characterization of Aerosol Samples. In most studies, separation and quantifi- 2. Experimental cation of PAHs in aerosol samples have been analyzed using a conventional GC-qMS [18]. Flame ionization detection 2.1. Air Sampling. /e total suspended particle (TSP) (FID) has also been widely used for quantification as it samples were collected at Asan Engineering Building, Ewha features a higher response to PAHs which contain only Womans University, Seoul, South Korea (37.56°N, 126.94°E, carbon and hydrogen, while oxygenates and other species 20 m above ground level), with a PUF sampler (Tisch, that contain heteroatoms tend to have a lower response TE-1000) on a quartz fiber filter (Quartz fiber filter, QFF, factor [19]. However, this nonspecific detector may not Ø10.16 cm, Whatman, UK). /e sampling site is located in distinguish inferences, which include a large fraction of the mixed resident area, commercial area, forest area, and aliphatic and aromatic compounds in aerosol samples from nearby roadside. A total of 67 filter samples were obtained alkylated PAH homologues. /e coupling of GC with MS is during summer (August 12–30, 2013) and winter (January increasingly becoming the analytical tool of choice in this 27–February 16, 2014) and day (9 a.m.∼6 p.m.) and night regard because of its superior selectivity and sensitivity. (8 p.m.∼6 a.m.). Prior to sampling, the quartz fiber filters Among the most common analyzers including TOF [20], ion were baked for 8 h in an electric oven at 550°C to remove trap, and qMS [21, 22], qMS is the most widely adopted possible organic contaminants. /e sampled filters were technique for routine analysis of PAHs [23]. GC-qMS data wrapped in aluminum foils and stored in a freezer at −20°C acquisition takes advantages of both a full mass scan range until analysis. (scan mode) and specific ion masses for target analytes (SIM mode). /e sensitivity in the SIM mode is higher than that in the scan mode of GC-qMS due to the increased dwell time 2.2. Chemicals. All organic solvents were of GC grade and on each monitored ion for trace analysis in some matrices purchased from Burdick and Jackson (Phillipsburg, NJ, such as in atmospheric aerosols [24, 25]. GC-TOFMS has USA). Standard solutions of target PAHs (Table 1 for their a much faster spectral acquisition rate than GC-qMS does, full chemical names and information) except Per and BeP which is up to 500 full mass scans per second [26]. Con- for quantitative analysis were purchased as a mixture at sequently, this system is able to widen the application of a concentration of 2000 µg/mL in dichloromethane from GC×GC techniques providing very narrow chromato- Supelco (Bellefonte, PA, USA). Per and BeP standards graphic peaks, typically 50∼600 ms at the baseline with Journal of Analytical Methods in Chemistry 3

Table 1: Information of target PAHs in the study. Retention time CAS Molecular Quantitative Qualifier Compound Abbreviation MW GC×GC-TOFMS number formula ion ion GC-qMS (min) tr1 (min) tr2 (s) a Naphthalene-d8 Nap-d8 1146-65-2 C10D8 136.2 136 137 12.25 13.40 1.34 Naphthalene Nap 91-20-3 C10H8 128.2 128 129 12.34 13.47 1.35 Acenaphthylene Acy 208-96-8 C12H8 152.2 152 153 19.08 19.56 1.55 a Acenaphthene-d10 Ace-d10 15067-26-2 C12D10 164.2 162 164 19.47 20.12 1.52 Acenaphthene Ace 83-32-9 C12H10 154.2 153 154 19.61 20.28 1.52 Fluorene F 86-73-7 C13H10 166.2 166 165 21.61 22.28 1.53 a Phenanthrene-d10 Phen-d10 1518-22-2 C14D10 188.2 188 189 25.91 25.96 1.68 Phenanthrene Phen 85-01-8 C14H10 178.2 178 179 26.01 26.04 1.71 Anthracene Ant 120-12-7 C14H10 178.2 178 179 26.14 26.20 1.68 a Fluoranthene-d10 Fla-d10 93951-69-0 C16D10 212.2 212 213 30.99 30.68 1.84 Fluoranthene Fla 206-44-0 C16H10 202.2 202 203 31.08 30.68 1.87 Pyrene Pyr 129-00-0 C16H10 202.2 202 203 32.25 31.56 1.98 Benz[a]anthracene BaA 56-55-3 C18H12 228.2 228 226 37.20 36.28 2.21 a Chrysene-d12 Chr-d12 1719-03-5 C18D12 240.3 240 236 37.42 36.28 2.27 Chrysene Chr 218-01-9 C18H12 228.3 228 226 37.54 36.44 2.26 Benzo[b]fluoranthene BbF 205-99-2 C20H12 252.3 252 253 41.55 40.12 2.76 Benzo[k]fluoranthene BkF 207-08-9 C20H12 252.3 252 253 41.65 40.28 2.74 Benzo[e]pyrene BeP 192-97-2 C20H12 252.3 252 253 42.90 41.08 3.16 Benzo[a]pyrene BaP 50-32-8 C20H12 252.3 252 253 43.09 41.24 3.23 a Perylene-d12 Per-d12 1520-96-3 C20D12 264.3 264 260 43.46 41.40 3.36 Perylene Per 198-55-0 C20H12 252.3 252 253 43.57 41.48 3.47 Indeno[1,2,3-cd] IP 193-39-5 C H 276.3 276 277 48.09 45.48 0.71 pyrene 22 12 Dibenz[a,h]anthracene DBahAnt 53-70-3 C22H14 278.3 278 279 48.12 45.64 0.82 Benzo[ghi] a BghiPer-d12 93951-66-7 C22D12 288.3 288 284 49.92 46.52 1.54 perylene-d12 Benzo[ghi]perylene BghiPer 191-24-2 C22H12 276.3 276 277 50.13 46.68 1.78 aInternal standard.

sufficient density of data points per chromatographic peak [27]. Environmental samples are generally complex, often Table 2: GC-qMS and GC×GC-TOFMS operating conditions. with more than hundreds of compounds containing structural isomers and homologues spread over a wide range × Parameters GC-qMS GC GC-TOFMS of concentration and volatility. Accordingly, multidimen- Injector settings sional separation is an advanced technique offering the Injection volume 1 μL 1 μL possibility of greatly enhanced selectivity using different Inlet mode Splitless Splitless separation mechanisms for the analysis of complex envi- Carrier gas He (99.999%) He (99.999%) Carrier gas flow 1.0 mL·min−1 1.3 mL·min−1 ronmental samples [28–30]. In this study, a set of columns Inlet temperature 280°C 300°C DB-5×DB-17 ms was applied to increase the resolution and GC oven temperature peak capacity. /e fast scanning Pegasus 4D TOFMS system Initial temperature 1 min at 60°C 1 min at 60°C was combined to allow efficient processing of data acqui- First rate 6°C/min to 310°C 6°C/min to 300°C sition, handling, peak detection, and deconvolution. In the Isothermal pause 15 min at 310°C 15 min at 300°C one-dimensional column, a 30 m-long DB-5 ms (5% 5°C, relative 2nd oven temperature diphenyl/95% dimethyl polysiloxane) stationary phase was — to the 2nd oven offset used to separate analytes based on volatility and combined temperature with a 1.17 m-long DB-17 ms column (50% diphenyl/50% Modulator dimethyl polysiloxane) allowing relative polarity-based 15°C, relative Modulator temperature × — to the 2nd separation. Figure 1 shows GC GC-TOFMS chromato- offset oven temperature grams of aerosol samples collected at day and night during Modulator period — 4.00 s winter in Seoul, South Korea. To compare the identification Hot pulse time — 1.00 s ability of GC×GC-TOFMS with GC-qMS, analysis with GC- Cool time between stages — 1.40 s qMS in the scan mode was performed. A comparison of the MS one-dimensional chromatograms of the same samples ob- Mass range 40∼550 40∼550 tained by GC-qMS is shown in Figure 2. 2D chromatograms Electron energy 70 eV 70 eV enable the visual classification of chemically related com- Ion source temperature 230°C 230°C pounds into groups. It was rare to see that the early-eluting 4 Journal of Analytical Methods in Chemistry

Masses: TIC 1.56172e + 006

2nd dimension rentention time (sec)

0 0 7:00.00 23:40.00 40:20.00

1st dimension retention time (min)

(a)

Masses: TIC 1.36252e + 006

2nd dimension rentention time (sec)

0 0 7:00.00 23:40.00 40:20.00

1st dimension retention time (min)

(b)

Figure 1: GC×GC-TOFMS plots of aerosol samples collected during day (a) and night (b) of winter in Seoul, Korea. A total of 251 and 297 peaks were identified in aerosol samples collected during day (a) and night (b), respectively. Aromatic and aliphatic classes were drawn to divide two regions for ease of viewing. analytes have an extreme volatility in the chromatogram, groups could be useful for source identification of atmo- as shown in Figure 2. Because of the large losses of these spheric aerosols by means of the large amount of chemical analytes during sample extraction and concentration, data handling. /e combined use with TOFMS provides particle-associated semivolatile analytes were mainly de- rapid and reliable identification of analytes using their tected and classified according to their aromatic and ali- deconvoluted pure mass spectra. /e major limitation of phatic hydrocarbon groups. qMS is its limited scan rate; therefore, quantification and Meanwhile, analytes from the GC-qMS chromatogram identification is seriously compromised because of the mass were separated based on their vapor pressures or boiling spectral skew due to the variations in ion abundances at points. /e GC×GC technique is rather well suited for group different regions of a chromatographic peak [31, 32]. /e separations, and classifying compounds into chemical-related numbers of identified chromatographic peaks analyzed by Journal of Analytical Methods in Chemistry 5

O OH O

15,00,000

10,00,000

Abundance 5,00,000 N N

0 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 Retention time (min)

(a)

15,00,000

10,00,000

Abundance 5,00,000

0 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 Retention time (min)

(b)

Figure 2: Total ion chromatograms of aerosol samples collected in day (a) and night (b) of winter in Seoul, Korea, obtained by GC-qMS. A total of 35 and 64 peaks were identiﬁed in aerosol samples collected during day (a) and night (b), respectively. /e analytes were separated based on their boiling points.

GC-qMS using the same signal threshold setting from the 3.2. Validation of GC-qMS and GC×GC-TOFMS for Quan- aerosol samples collected at day and night were 35 and 64, tification of PAHs. GC-qMS and GC×GC-TOFMS were respectively. In the case of results obtained by GC×GC- tested individually in order to evaluate their analytical TOFMS, 251 and 297 peaks from the day- and night-time performances. /e calibration linearity (regression co- aerosol samples were, respectively, assigned by individual efficient, R2) and relative response factor (RRF) are pre- spectral deconvolution. As a result, phthalic anhydride and sented in Table 3. /e RRF is the ratio between a signal 1,2-naphthalic anhydride as the markers of secondary for- produced by an individual native analyte and the corre- mation for gas-phase PAH reactions were identified in the sponding isotopically labeled analogue of the analyte (as an aerosol sample, as shown in Figure 3. Since the products internal standard). For calculating RRF, 2 ng of each target formed through photochemical reactions are often more PAH and each corresponding deuterated internal standard toxic than their parent PAHs in atmosphere [17], sig- was spiked, and the relative sensitivity in both the methods nificant efforts have been expended to identify the pho- was compared. Despite the high-speed scanning perfor- tochemical products with PAHs in the fields of mance of GC×GC-TOFMS, the RRFs obtained by this atmospheric or environmental sciences. In the case of method were approximately equivalent to those obtained by results obtained using GC-qMS, phthalic anhydride and GC-qMS. RRF expresses the sensitivity of a detector for 1,2-naphthalic anhydride were not detected in the same a given substance relative to a standard substance [35, 36]. sample. Limitations of one-dimensional separation have /us, it indicated that the sensitivity of GC×GC-TOFMS been reported for these photochemical products and relative to target PAHs is comparable in quantitative complex mixtures of the aerosol sample because of their analysis. Calibration curves were generated using the peak diverse polarities in a single run [33, 34]. Contrastively, area for the 18 PAHs at seven concentrations ranging from two anhydrides associated with secondary organic aerosol 0.01 to 10 μg/mL. /e linearity was assessed by calculating formation were clearly separated and detected by GC×GC- the regression equation and the correlation coefficient by the TOFMS. /erefore, it showed advantages for nontarget least squares method, as shown in Table 3. /e R2 values screening to identify molecular markers or chemical were greater than 0.999 for GC-qMS and 0.99 for GC×GC- patterns more representative of the aerosol state observed TOFMS. Although data processing for quantification by in ambient air. GC×GC-TOFMS was derived from the combined peak areas 6 Journal of Analytical Methods in Chemistry

Masses: 104 126 178 200000

2.355

1.855

1.355 2nd dimension rentention time (sec) rentention 2nd dimension

0.855 0 943.2 1143.2 1343.2 1543.2

1st dimension retention time (min)

Figure 3: GC×GC chromatograms and mass spectrums of phthalic anhydride (marked as green) and 1,2-naphthalic anhydride (marked as yellow) in the aerosol sample. GC×GC chromatograms of phthalic anhydride and 1,2-naphthalic anhydride were certiﬁed by molecular ions of m/z 148 and 198, respectively.

Table 3: Relative response factors (RRFs) and calibrations of 18 PAHs obtained by the compared methods. GC-qMS GC×GC-TOFMS Compound RRFa Slope Intercept R2 RRF Slope Intercept R2 Nap 1.04 0.515 −0.003 0.9999 1.69 0.560 0.049 0.9971 Acy 1.57 0.808 −0.004 1.0000 1.95 1.026 −0.012 0.9999 Ace 1.03 0.439 0.009 0.9999 1.16 0.553 −0.006 0.9994 F 1.29 0.667 −0.006 1.0000 1.11 0.609 −0.021 0.9997 Phe 1.17 0.572 −0.004 0.9998 1.47 0.763 −0.039 0.9979 Ant 0.98 0.547 −0.017 0.9992 0.93 0.431 −0.010 0.9982 Fla 1.30 0.678 −0.001 1.0000 1.43 0.802 −0.021 0.9995 Pyr 1.31 0.686 −0.004 0.9999 1.62 0.910 −0.055 0.9972 BaA 0.98 0.575 −0.019 0.9997 1.42 0.574 −0.004 0.9998 Chr 1.06 0.563 −0.003 1.0000 1.26 0.651 −0.005 0.9998 BbF 0.99 0.535 −0.008 0.9999 1.69 0.848 −0.028 0.9993 BkF 1.11 0.576 −0.011 0.9998 0.88 0.327 −0.012 0.9977 BeP 0.91 0.455 −0.007 0.9996 0.90 0.530 −0.014 0.9997 BaP 0.88 0.505 −0.014 0.9997 0.83 0.477 −0.030 0.9985 Per 0.89 0.475 −0.008 0.9998 1.11 0.521 -0.023 0.9979 IP 1.37 0.717 −0.023 0.9995 1.25 0.660 -0.062 0.9922 DBahAnt 1.24 0.629 −0.019 0.9996 1.16 0.490 -0.074 0.9898 BghiPer 1.24 0.594 −0.010 1.000 1.53 0.710 -0.036 0.9991 a RRF expresses the sensitivity of a detector for a given analyte relative to its corresponding deuterated internal standards; RRF � (AxCis)/(AisCx), where Ax is the peak area of a quantifying ion for a given analyte being measured; Ais is the peak area of a quantifying ion for its corresponding internal standard; Cx is the concentration of a given analyte; and Cis is the concentration of its corresponding internal standard. for the slices of modulated peaks in contrast to production of Naturally, the development of quantitative GC×GC studies the single measured peak by GC-qMS, the results meet the based on the quantitative results associated with sophisti- criteria for acceptable linearity within this calibration range. cated implementation for modulated peaks has been delayed Journal of Analytical Methods in Chemistry 7

Table 4: Limits of detection and quantification and recoveries of 18 PAHs obtained by the compared methods. LODa (ng) LOQb (ng) Recovery ± RSD (%) Compound GC-qMS GC×GC-TOFMS GC-qMS GC×GC-TOFMS GC-qMS GC×GC-TOFMS Nap 0.07 0.40 0.21 1.19 94.4 ± 4.2 135 ± 45 Acy 0.17 0.07 0.51 0.22 119 ± 12 116 ± 15 Ace 0.05 0.17 0.16 0.52 105 ± 5.3 105 ± 7.8 F 0.04 0.15 0.13 0.44 158 ± 28 130 ± 29 Phe 0.10 0.34 0.31 1.03 94.5 ± 5.3 86.3 ± 16 Ant 0.19 0.31 0.58 0.92 90.4 ± 4.6 95.1 ± 20 Fla 0.05 0.14 0.16 0.41 90.3 ± 3.9 105 ± 13 Pyr 0.08 0.36 0.25 1.09 97.4 ± 5.3 97.2 ± 13 BaA 0.12 0.09 0.37 0.27 93.4 ± 4.9 86.9 ± 8.2 Chr 0.04 0.08 0.13 0.24 95.8 ± 5.8 101 ± 16 BbF 0.05 0.18 0.15 0.53 96.1 ± 5.7 92.3 ± 10 BkF 0.09 0.35 0.28 1.05 94.2 ± 6.5 105 ± 12 BeP 0.13 0.13 0.40 0.38 92.6 ± 5.8 92.7 ± 5.7 BaP 0.12 0.24 0.37 0.72 93.6 ± 5.3 104 ± 9.0 Per 0.11 0.34 0.32 1.02 93.0 ± 5.5 92.5 ± 8.6 IP 0.15 0.65 0.16 1.94 95.0 ± 5.4 93.9 ± 8.5 DBahAnt 0.13 1.05 0.40 3.14 94.9 ± 5.5 95.8 ± 5.7 BghiPer 0.09 0.22 0.27 0.66 94.6 ± 6.0 87.0 ± 8.5 aLOD, smallest amount of analyte that is statistically different from the blank; bLOQ, smallest amount of analyte that can be measured with reasonable accuracy. compared with qualitative reports. Recently, the approach to curve (S) as LOD � 3.3 × (s/S) and LOQ � 10 × (s/S). /e quantifying multiple analytes at once with comprehensive LOD and LOQ for each PAH compound obtained from both two-dimensional GC has been extensively studied in ac- the methods are shown in Table 4. /e LOD and LOQ values cordance with the improvement of data processing for the of the 18 PAH compounds obtained by GC-qMS were integration of modulated peaks [37, 38]. In this study, the similar to the results of previous studies [10, 41, 42]. /us, modulated peaks of each PAH was automatically combined the suitability of GC×GC-TOFMS for quantification of and integrated by the ChromaTOF software based on PAHs was proven by comparing the results with those a similarity of spectra within an allowable time difference obtained using GC-qMS. between the second dimension peaks in the neighboring slices of the chromatogram. Recovery test was performed by 4. Conclusion spiking known amounts of the 18 PAH compounds in a prebaked clean filter at a final concentration of 2 μg/mL A fast scanning GC×GC-TOFMS was compared to a GC-qMS and analyses of each through all the experiment procedures for the determination of PAHs in aerosol samples. For sep- were compared using the two different methods. Six du- aration, identification, and characterization, GC×GC-TOFMS plicate tests were performed, and the results of the recovery was advantageous over GC-qMS owing to the increased peak are shown in Table 4. /e average recoveries were in the capacity, and its results showed enhanced detectability and range of 90.3 to 158% with relative standard deviations structured chromatograms for nontarget analysis. /e quali- (RSDs) ranging from 3.9 to 28% for GC-qMS, while the tative mass separation by TOFMS combined with an auto- recoveries were from 86.3 to 135% for GC×GC-TOFMS, mated peak-finding capability provided the resolution of with RSDs ranging from 5.7 to 45%. Most of the targeted complex mixed mass spectra, resulting from overlapping PAH compounds were afforded acceptable recoveries, ex- chromatographic peaks and spectral deconvolution of indi- cluding F and Nap by using the two analytical methods due vidual mass spectra for unknown analytes. Furthermore, the to the high volatility of these compounds. Compared with obtained quantitative results such as LODs, LOQs, and re- the reproducibility as expressed in %RSDs, the values ob- coveries of the 18 target PAHs were approximately equivalent tained by GC-qMS were slightly lower than those obtained for both the analytical methods. /us, GC×GC-TOFMS had by GC×GC-TOFMS; however, the %RSD values of the advantages for the simultaneous quantification and qualifi- targeted PAHs excluding F and Nap were acceptable (<20% cation of PAHs and other organic compounds in a single run. RSD). /ese observations may vary for the versatile GC×GC Because of its high degree of separation and capability of technique, since the reproducibility of the modulation phase spectral deconvolution of overlapping peaks in highly complex is dependent on the type of modulator, the stability of the samples, comprehensive GC×GC-TOFMS may become stationary phases, and the chemistry of the analyte, re- a useful platform in many other fields of research. garding interaction with the stationary phase as presented in several prior studies [39, 40]. /e LOD and LOQ were Conflicts of Interest determined based on the standard deviation (SD) of the intersection of the analytical curve (s) and the slope of the /e authors declare that they have no conflicts of interest. 8 Journal of Analytical Methods in Chemistry

Acknowledgments products of gas phase PAHs in PM2.5,” Atmospheric Envi- ronment, vol. 55, pp. 17–25, 2012. /is research was supported by the Bio-Synergy Research [14] R. J. Vreuls, J. Dallüge, and U. A. T. Brinkman, “Gas Project (no. NRF-2017M3A9C4065961) of the Ministry of chromatography–time-of-flight mass spectrometry for sen- Science, ICT, and Future Planning through the National sitive determination of organic microcontaminants,” Journal Research Foundation and the Korea Basic Science Institute of Microcolumn Separations, vol. 11, no. 9, pp. 663–675, 1999. Grant (no. C37705). /is research was also supported by the [15] C. Weickhardt, F. Moritz, and J. Grotemeyer, “Time-of-flight Basic Science Research Program through the National Re- mass spectrometry: state-of the-art in chemical analysis and molecular science,” Mass Spectrometry Reviews, vol. 15, no. 3, search Foundation of Korea (NRF) funded by the Ministry of pp. 139–162, 1996. Education (no. NRF-2016R1A2B4015143) [16] M. X. Xie, F. Xie, Z. W. Deng, and G. S. Zhuang, “De- termination of polynuclear aromatic hydrocarbons in aerosol References by solid-phase extraction and gas chromatography–mass spectrum,” Talanta, vol. 60, no. 6, pp. 1245–1257, 2003. [1] IARC, “Polynuclear aromatic compounds, Part 1. Chemical, [17] R. Atkinson and J. Arey, “Atmospheric chemistry of gas-phase environmental and experimental data,” IARC Monographs on polycyclic aromatic hydrocarbons: formation of atmospheric the Evaluation of the Carcinogenic Risk of Chemicals to mutagens,” Environmental Health Perspectives, vol. 102, no. 4, Humans, vol. 32, pp. 1–453, 1983. pp. 117–126, 1994. [2] World Health Organization, “Environmental health criteria,” [18] J. D. Pleil, T. L. Vossler, W. A. McClenny, and K. D. Oliver, in International Programme on Chemical Safety (IPCS), “Optimizing sensitivity of SIM mode of GC/MS analysis for vol. 171, WHO, Geneva, Switzerland, 1998. EPA’s TO-14 air toxics method,” Journal of the Air & Waste [3] V. Vestreng, “Emission data reported to UNECE/EMEP: Management Association, vol. 41, no. 3, pp. 287–293, 1991. quality assurance and trend analysis and presentation of [19] D. L. Poster, M. M. Schantz, L. C. Sander, and S. A. Wise, WebDab: MSC-W status report 2002,” Research report, “Analysis of polycyclic aromatic hydrocarbons (PAHs) in University of Oslo, Oslo, Norway, 2002. environmental samples: a critical review of gas chromato- [4] L.-B. Liu, L. Yan, J.-M. Lin, T. Ning, K. Hayakawa, and graphic (GC) methods,” Analytical and Bioanalytical T. Maeda, “Development of analytical methods for polycyclic Chemistry, vol. 386, no. 4, pp. 859–881, 2006. aromatic hydrocarbons (PAHs) in airborne particulates: [20] W. Welthagen, J. Schnelle-Kreis, and R. Zimmermann, a review,” Journal of Environmental Sciences, vol. 19, no. 1, “Search criteria and rules for comprehensive two-dimensional pp. 1–11, 2007. gas chromatography–time-of-flight mass spectrometry anal- [5] E. Cavalieri, R. Roth, E. Rogan, C. Grandjean, and J. Althoff, ysis of airborne particulate matter,” Journal of Chromatog- “Mechanisms of tumor initiation by polycyclic aromatic raphy A, vol. 1019, no. 1-2, pp. 33–249, 2003. [21] A. Filipkowska, L. Lubecki, and G. Kowalewska, “Polycyclic hydrocarbons,” Carcinogenesis, vol. 3, pp. 273–287, 1978. aromatic hydrocarbon analysis in different matrices of the [6] Z. Zelinkova and T. Wenzl, “/e occurrence of 16 EPA PAHs marine environment,” Analytica Chimica Acta, vol. 547, no. 2, in food—a review,” Polycyclic Aromatic Compounds, vol. 35, pp. 243–254, 2005. no. 2–4, pp. 248–284, 2015. [22] K. Ravindra, A. F. L. Godoi, L. Bencs, and R. Van Grieken, [7] T. Wenzl, R. Simon, E. Anklam, and J. Kleiner, “Analytical “Low-pressure gas chromatography–ion trap mass spec- methods for polycyclic aromatic hydrocarbons (PAHs) in trometry for the fast determination of polycyclic aromatic food and the environment needed for new food legislation in hydrocarbons in air samples,” Journal of Chromatography A, the European Union,” Trends in Analytical Chemistry, vol. 25, vol. 1114, no. 2, pp. 278–281, 2006. pp. 716–725, 2006. [23] M. Bergknut, K. Frech, P. L. Andersson, P. Haglund, and [8] H. I. Abdel-Shafy and M. S. Mansour, “A review on polycyclic M. Tysklind, “Characterization and classification of complex aromatic hydrocarbons: source, environmental impact, effect PAH samples using GC–qMS and GC–TOFMS,” Chemo- on human health and remediation,” Egyptian Journal of Pe- sphere, vol. 65, no. 11, pp. 2208–2215, 2006. troleum, vol. 25, no. 1, pp. 107–123, 2016. [24] T. Tran, Characterization of Crude Oils and Atmospheric [9] U. Poschl,¨ “Atmospheric aerosols: composition, trans- Organic Compounds by Using Comprehensive Two- formation, climate and health effects,” Angewandte Chemie Dimensional Gas Chromatography Technique (GC×GC), International Edition, vol. 44, no. 46, pp. 7520–7540, 2005. Ph.D. thesis, Applied Sciences, RMIT University, Melbourne [10] M. A. Bari, G. Baumbach, B. Kuch, and G. Scheffknecht, VIC, Australia, 2009. “Particle-phase concentrations of polycyclic aromatic hy- [25] Environmental Protection Agency (EPA), Compendium Method drocarbons in ambient air of rural residential areas in TO-13A, Determination of Polycyclic Aromatic Hydrocarbons southern Germany,” Air Quality, Atmospheric Health, vol. 3, (PAHs) in Ambient Air Using Gas Chromatography/Mass no. 2, pp. 103–116, 2010. Spectrometry (GC/MS), EPA, Cincinnati, OH, USA, 1999. [11] D. A. Lane, A. Leithead, M. Baroi, J. Y. Lee, and L. A. Graham, [26] S. H. Jeon, J. H. Shin, Y. P. Kim, and Y. G. Ahn, “De- “/e detection of polycyclic aromatic compounds in air termination of volatile alkylpyrazines in microbial samples samples by GC×GC-TOFMS,” Polycyclic Aromatic Com- using gas chromatography-mass spectrometry coupled with pounds, vol. 28, no. 4-5, pp. 545–561, 2008. head space-solid phase microextraction,” Journal of Analytical [12] D. A. Lane and J. Y. Lee, “Detection of known photochemical Science and Technology, vol. 7, no. 1, p. 16, 2016. decomposition products of PAH in particulate matter from [27] A. R. Fernández-Alba, TOF-MS within Food and Environ- pollution episodes in Seoul, Korea,” Polycyclic Aromatic mental Analysis, vol. 58, Elsevier, Amsterdam, Netherlands, Compounds, vol. 30, no. 5, pp. 309–320, 2010. 2012. [13] J. Y. Lee, D. A. Lane, J. B. Heo, S.-M. Yi, and Y. P. Kim, [28] J. H. Winnike, X. Wei, K. J. Knagge, S. D. Colman, “Quantification and seasonal pattern of atmospheric reaction S. G. Gregory, and X. Zhang, “Comparison of GC-MS and Journal of Analytical Methods in Chemistry 9

GC×GC-MS in the analysis of human serum samples for [42] H. C. Menezes and Z. de Lourdes Cardeal, “Determination of biomarker discovery,” Journal of Proteome Research, vol. 14, polycyclic aromatic hydrocarbons from ambient air partic- no. 4, pp. 1810–1817, 2015. ulate matter using a cold fiber solid phase microextraction gas [29] L. I. Osemwengie and G. W. Sovocool, “Evaluation of chromatography–mass spectrometry method,” Journal of comprehensive 2D gas chromatography-time-of-flight mass Chromatography A, vol. 1218, no. 21, pp. 3300–3305, 2011. spectrometry for 209 chlorinated biphenyl congeners in two chromatographic runs,” Chromatography Research In- ternational, vol. 2011, Article ID 675920, 14 pages, 2011. [30] J. Zrostlıkov́ a,´ J. Hajˇslova,´ and T. Cajka,ˇ “Evaluation of two- dimensional gas chromatography–time-of-flight mass spectrometry for the determination of multiple pesticide residues in fruit,” Journal of Chromatography A, vol. 1019, no. 1-2, pp. 173–186, 2003. [31] P. Antle, C. D. Zeigler, Y. Gankin, and J. A. Robbat, “New spectral deconvolution algorithms for the analysis of polycyclic aromatic hydrocarbons and sulfur heterocycles by comprehensive two-dimensional gas chromatography- quadrupole mass spectrometery,” Analytical Chemistry, vol. 85, no. 21, pp. 10369–10376, 2013. [32] S. Samanipour, P. Dimitriou-Christidis, J. Gros, A. Grange, and J. Samuel Arey, “Analyte quantification with comprehensive two-dimensional gas chromatography: assessment of methods for baseline correction, peak delineation, and matrix effect elimination for real samples,” Journal of Chromatog- raphy A, vol. 1375, pp. 123–139, 2015. [33] P. Mills and W. Guise Jr., “A multidimensional gas chromatographic method for analysis of n-butane oxidation reaction products,” Journal of Chromatographic Science, vol. 34, no. 10, pp. 431–459, 1996. [34] R. M. Flores and P. V. Doskey, “Using multidimensional gas chromatography to group secondary organic aerosol species by functionality,” Atmospheric Environment, vol. 96, pp. 310–321, 2014. [35] S. Pongpiachan, P. Hirunyatrakul, I. Kittikoon, and C. Khumsup, “Parameters influencing on sensitivities of polycyclic aromatic hydrocarbons measured by Shimadzu GCMS-QP2010 ultra,” in Advanced Gas Chromatography– Progress in Agricultural, Biomedical and Industrial Applica- tions, M. Ali Mohd, Ed., InTech, Rijeka, Croatia, 2012. [36] European Pharmacopoeia 7.0, Section 2.2.46, Chromato- graphic Separation Techniques, 2010. [37] P. Marriott and C. Mühlen,“/e modulation ratio in comprehensive two-dimensional gas chromatography: a review of fundamental and practical considerations,” Scientia Chro- matographica, vol. 8, no. 1, pp. 7–23, 2016. [38] J. Krupcik, P. Majek, R. Gorovenko, J. Blasko, R. Kubinec, and P. Sandra, “Considerations on the determination of the limit of detection and the limit of quantification in one- dimensional and comprehensive two-dimensional gas chromatography,” Journal of Chromatography A, vol. 1396, no. 117, pp. 117–130, 2015. [39] P. M. Antle, C. D. Zeigler, N. M. Wilton, and A. Robbat Jr., “A more accurate analysis of alkylated PAH and PASH and its implications in environmental forensics,” International Journal of Environmental Analytical Chemistry, vol. 94, no. 4, pp. 332–347, 2014. [40] T. Cajka, “Gas chromatography–time-of-flight mass spectrometry in food and environmental analysis,” in Compre- hensive Analytical Chemistry, I. Ferrer, Ed., pp. 271–302, Elsevier, Amsterdam, Netherlands, 2013. [41] B. Lazarov, R. Swinnen, M. Spruyt et al., “Optimisation steps of an innovative air sampling method for semi volatile organic compounds,” Atmospheric Environment, vol. 79, pp. 780–786, 2013.